id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 3
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-12-13 00:00:00
| authors_parsed
listlengths 1
3.44k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2512.10003
|
Kaitlin Webber
|
Kaitlin B. Webber, Terese T. Hansen, Jennifer L. Marshall, Alexander P. Ji, Ting S. Li, Gary S. Da Costa, Lara R. Cullinane, Denis Erkal, Sergey E. Koposov, Kyler Kuehn, Geraint F. Lewis, Dougal Mackey, Sarah L. Martell, Andrew B. Pace, Nora Shipp, Jeffrey D. Simpson, Zhen Wan, Daniel B. Zucker, Victor A. Alvarado, Joss Bland-Hawthorn, Guilherme Limberg, Gustavo E. Medina, and Sam A. Usman
|
Detailed Chemical Abundance Analysis of the Brightest Stars in the Turranburra and Willka Yaku Stellar Streams
|
21 pages, 10 tables, 6 figures
| null | null | null |
astro-ph.SR astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a detailed chemical abundance analysis of the three brightest known stars from each of the Turranburra and Willka Yaku stellar streams using high-resolution Magellan/MIKE spectra. Abundances for 27 elements, ranging from carbon to dysprosium, were derived. Our results support the original classification that Turranburra, with a low average metallicity of $\mathrm{[Fe/H]=-2.45} \pm 0.07$, likely originates from a dwarf-galaxy progenitor. Willka Yaku has a low average metallicity of $\mathrm{[Fe/H]=-2.35 \pm 0.03}$ with a small scatter in the abundances, consistent with a globular cluster progenitor as suggested by previous studies. Both streams exhibit mild enhancements in neutron-capture elements, with averages of $\mathrm{[Eu II/Fe]}=$ $0.47 \pm{0.09}$ for Turranburra and $0.44 \pm{0.05}$ for Willka Yaku, consistent with enrichment from an $r$-process event. A similar enrichment is observed in other stellar streams, and we further discuss this signature as it relates to the potential enrichment histories of these two streams.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:00:14 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Webber",
"Kaitlin B.",
""
],
[
"Hansen",
"Terese T.",
""
],
[
"Marshall",
"Jennifer L.",
""
],
[
"Ji",
"Alexander P.",
""
],
[
"Li",
"Ting S.",
""
],
[
"Da Costa",
"Gary S.",
""
],
[
"Cullinane",
"Lara R.",
""
],
[
"Erkal",
"Denis",
""
],
[
"Koposov",
"Sergey E.",
""
],
[
"Kuehn",
"Kyler",
""
],
[
"Lewis",
"Geraint F.",
""
],
[
"Mackey",
"Dougal",
""
],
[
"Martell",
"Sarah L.",
""
],
[
"Pace",
"Andrew B.",
""
],
[
"Shipp",
"Nora",
""
],
[
"Simpson",
"Jeffrey D.",
""
],
[
"Wan",
"Zhen",
""
],
[
"Zucker",
"Daniel B.",
""
],
[
"Alvarado",
"Victor A.",
""
],
[
"Bland-Hawthorn",
"Joss",
""
],
[
"Limberg",
"Guilherme",
""
],
[
"Medina",
"Gustavo E.",
""
],
[
"Usman",
"Sam A.",
""
]
] |
2512.10004
|
Sha Li
|
Sha Li, Ayush Sadekar, Nathan Self, Yiqi Su, Lars Andersland, Mira Chaplin, Annabel Zhang, Hyoju Yang, James B Henderson, Krista Wigginton, Linsey Marr, T.M. Murali, Naren Ramakrishnan
|
Exploring LLMs for Scientific Information Extraction Using The SciEx Framework
| null | null | null | null |
cs.AI cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language models (LLMs) are increasingly touted as powerful tools for automating scientific information extraction. However, existing methods and tools often struggle with the realities of scientific literature: long-context documents, multi-modal content, and reconciling varied and inconsistent fine-grained information across multiple publications into standardized formats. These challenges are further compounded when the desired data schema or extraction ontology changes rapidly, making it difficult to re-architect or fine-tune existing systems. We present SciEx, a modular and composable framework that decouples key components including PDF parsing, multi-modal retrieval, extraction, and aggregation. This design streamlines on-demand data extraction while enabling extensibility and flexible integration of new models, prompting strategies, and reasoning mechanisms. We evaluate SciEx on datasets spanning three scientific topics for its ability to extract fine-grained information accurately and consistently. Our findings provide practical insights into both the strengths and limitations of current LLM-based pipelines.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:00:20 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Li",
"Sha",
""
],
[
"Sadekar",
"Ayush",
""
],
[
"Self",
"Nathan",
""
],
[
"Su",
"Yiqi",
""
],
[
"Andersland",
"Lars",
""
],
[
"Chaplin",
"Mira",
""
],
[
"Zhang",
"Annabel",
""
],
[
"Yang",
"Hyoju",
""
],
[
"Henderson",
"James B",
""
],
[
"Wigginton",
"Krista",
""
],
[
"Marr",
"Linsey",
""
],
[
"Murali",
"T. M.",
""
],
[
"Ramakrishnan",
"Naren",
""
]
] |
2512.10005
|
Parvinder Solanki
|
Parvinder Solanki, Igor Lesanovsky, and Gabriele Perfetto
|
Universal relaxation speedup in open quantum systems through transient conditional and unconditional resetting
| null | null | null | null |
cond-mat.stat-mech quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Speeding up the relaxation dynamics of many-body quantum systems is important in a variety of contexts, including quantum computation and state preparation. We demonstrate that such acceleration can be universally achieved via transient stochastic resetting. This means that during an initial time interval of finite duration, the dynamics is interrupted by resets that take the system to a designated state at randomly selected times. We illustrate this idea for few-body open systems and also for a challenging many-body case, where a first-order phase transition leads to a divergence of relaxation time. In all scenarios, a significant and sometimes even exponential acceleration in reaching the stationary state is observed, similar to the so-called Mpemba effect. The universal nature of this speedup lies in the fact that the design of the resetting protocol only requires knowledge of a few macroscopic properties of the target state, such as the order parameter of the phase transition, while it does not necessitate any fine-tuned manipulation of the initial state.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:00:28 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Solanki",
"Parvinder",
""
],
[
"Lesanovsky",
"Igor",
""
],
[
"Perfetto",
"Gabriele",
""
]
] |
2512.10006
|
Andrea Conti
|
Andrea Conti, Yolanda Lozano, Christopher Rosen
|
Monodromy Defects in Massive Type IIA
|
38 pages, 2 figues, 1 Appendix
| null | null | null |
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper we study solutions to massive Type IIA supergravity which we propose are dual to co-dimension 2 monodromy defects in 6d (1,0) CFTs realised in NS5-D6-D8 brane systems, as well as in the 5d Sp(N) fixed point theory. In the first case the defects are studied holographically as solutions to 7d $U(1)$ gauged supergravity that asymptote locally to its maximally supersymmetric $\text{AdS}_7$ vacuum away from the defects. In the second case they are dual to solutions to 6d $U(1)^2$ gauged supergravity that asymptote to its maximally supersymmetric $\text{AdS}_6$ vacuum. These solutions are then uplifted to massive Type IIA supergravity using known consistent truncations previously constructed in the literature. We compute the defect entanglement entropy and provide evidence that for the 3d defects, the entanglement entropy can be written as a linear combination of the free energy and conformal weight of the defect. Finally, we construct new co-dimension 2 monodromy defects in 6d and 5d CFTs compactified on Riemann surfaces and/or spindles.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:00:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Conti",
"Andrea",
""
],
[
"Lozano",
"Yolanda",
""
],
[
"Rosen",
"Christopher",
""
]
] |
2512.10007
|
Michael Greklek-McKeon
|
Michael Greklek-McKeon, Jonathan Gomez Barrientos, Heather A. Knutson, Sebasti\'an Z\'u\~niga-Fern\'andez, Francisco J. Pozuelos, Morgan Saidel, W. Garrett Levine, Renyu Hu, Fei Dai, Tansu Daylan, John P. Doty, David R. Rodriguez, Joseph D. Twicken, David W. Latham, Jon M. Jenkins, Richard P. Schwarz
|
Validation of a Third Earth-sized Planet in the TOI-2267 Binary System
|
Accepted for publication in The Astronomical Journal. 14 pages, 4 figures, 2 tables. Light curve data is available in the arXiv source files
| null | null | null |
astro-ph.EP
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We report the validation of a third terrestrial exoplanet in the nearby (22 pc) TOI-2267 system. TOI-2267 is a binary system with stellar components TOI-2267A (M5, 3030 K) and TOI-2267B (M6, 2930 K), with an on-sky separation of 0.$^{\prime\prime}$384 (8 au projected separation). TOI-2267 hosts two Earth-sized planets (TOI-2267 b, $1.00\pm0.11 R_{\oplus}$, and TOI-2267 c, $1.14\pm0.13 R_{\oplus}$, if orbiting the primary star; or $1.22\pm0.29 R_{\oplus}$ and $1.36\pm0.33 R_{\oplus}$, respectively, if orbiting the secondary star) with orbital periods of 2.3 and 3.5 days. This system also contains a third Earth-sized planet candidate with an orbital period of 2.0 days that was previously identified as a likely planet with a low false-positive probability, but could not be firmly validated due to the lack of independent observations beyond TESS data. We combine two new transit observations from the 5.1m Hale Telescope at Palomar Observatory with archival TESS data and high-resolution imaging to statistically validate the planetary nature of TOI-2267 d ($0.98\pm0.09 R_{\oplus}$ if orbiting the primary star, or $1.77\pm0.43 R_{\oplus}$ if orbiting the secondary star) using the updated TRICERATOPS+ pipeline. We attempt to determine the host star for TOI-2267 d using transit shape stellar density analysis, but are unable to conclusively assign a host. Our validation of TOI-2267 d suggests that TOI-2267 is either the first known double transiting M dwarf binary system, or hosts three planets in an extremely compact orbital configuration.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:00:39 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Greklek-McKeon",
"Michael",
""
],
[
"Barrientos",
"Jonathan Gomez",
""
],
[
"Knutson",
"Heather A.",
""
],
[
"Zúñiga-Fernández",
"Sebastián",
""
],
[
"Pozuelos",
"Francisco J.",
""
],
[
"Saidel",
"Morgan",
""
],
[
"Levine",
"W. Garrett",
""
],
[
"Hu",
"Renyu",
""
],
[
"Dai",
"Fei",
""
],
[
"Daylan",
"Tansu",
""
],
[
"Doty",
"John P.",
""
],
[
"Rodriguez",
"David R.",
""
],
[
"Twicken",
"Joseph D.",
""
],
[
"Latham",
"David W.",
""
],
[
"Jenkins",
"Jon M.",
""
],
[
"Schwarz",
"Richard P.",
""
]
] |
2512.10008
|
Maxime Gadioux
|
Maxime Gadioux, Harvey S. Reall, Jorge E. Santos
|
Formation of extremal Reissner-Nordstr\"om black holes: insights from numerics
|
22 pages, 12 figures
| null | null | null |
gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
An extremal Reissner-Nordstr\"om black hole can form in finite time in the gravitational collapse of a massless charged scalar field. The proof of this is based on the method of characteristic gluing, which involves making an Ansatz for the scalar field at the horizon. We perform a numerical investigation of the characteristic gluing procedure for several different Ans\"atze. In each case, gluing is possible only if the final black hole mass is large enough. We find that the minimum required mass varies significantly for different Ans\"atze. We also consider the effect of including a mass term for the scalar field. In this case, for each Ansatz we determine the maximum mass-to-charge ratio for the scalar field such that gluing is possible. Analogous results are obtained for a non-zero cosmological constant.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:00:49 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gadioux",
"Maxime",
""
],
[
"Reall",
"Harvey S.",
""
],
[
"Santos",
"Jorge E.",
""
]
] |
2512.10009
|
Berke Ricketti
|
Berke Vow Ricketti, Victoria Yankelevich, Chris Benson, Renske Smit, Sebastian Kamann, Ettore Pedretti, Sebastian Marino, Gerard van Belle, Stephen Eales, Chris Bee, Mark Wyatt, Matthew Smith, Tim D. Pearce, Emily Williams, Rebecca Harwin, David Pearson, Andy Vick, Giorgio Savini, Taro Matsuo, Hiroshi Matsuo, Locke Spencer, and David T. Leisawitz
|
The SPace-based InterFerometer Feasibility (SPIFF) Project: Enabling Future High-Resolution Astronomy Across the EM Spectrum
| null | null | null | null |
astro-ph.IM
|
http://creativecommons.org/licenses/by/4.0/
|
A plethora of astronomical science cases can only be achieved with high angular resolution observations, and we can expect the number of these to grow as astronomers are constrained by the size limitations of single-aperture space telescopes, making space-based interferometry inevitable. However, the enabling technologies do not have flight heritage at the system level, and the concept remains immature to much of the community, meaning no direct-detection synthetic-aperture space-based interferometer has yet flown and an opportunity exists for the UK to take a world leading role. Here we propose the SPace-based InterFerometry Feasibility (SPIFF) Project as a program to address both issues simultaneously by: 1) completing a thorough survey of the science cases across the EM spectrum that would directly benefit from, or be impossible without, space-based interferometry; 2) down selecting key requirements via a Science Traceability Matrix mapping exercise; 3) host a workshop for the UK astronomical community to consolidate these findings; 4) build a technology demonstration mission to raise TRL and achieve flight heritage of critical technologies. Such a program positions the UK as the partner of choice for any future NASA or ESA space-based interferometry mission, allowing the UK to lead groundbreaking scientific discoveries, while also directly benefiting the UK industrial base by advancing domestic exportable technologies and building direct synergy with other UK space priorities. Indeed, the UK is uniquely positioned to lead in space-based interferometry, possessing a rare trifecta of domestic strengths: world-class expertise in ground-based interferometry and space-based instrumentation; commercial entities developing mission-critical technologies; and scientists whose research spans the full range of science cases that would benefit directly from space-based interferometry.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:01:00 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ricketti",
"Berke Vow",
""
],
[
"Yankelevich",
"Victoria",
""
],
[
"Benson",
"Chris",
""
],
[
"Smit",
"Renske",
""
],
[
"Kamann",
"Sebastian",
""
],
[
"Pedretti",
"Ettore",
""
],
[
"Marino",
"Sebastian",
""
],
[
"van Belle",
"Gerard",
""
],
[
"Eales",
"Stephen",
""
],
[
"Bee",
"Chris",
""
],
[
"Wyatt",
"Mark",
""
],
[
"Smith",
"Matthew",
""
],
[
"Pearce",
"Tim D.",
""
],
[
"Williams",
"Emily",
""
],
[
"Harwin",
"Rebecca",
""
],
[
"Pearson",
"David",
""
],
[
"Vick",
"Andy",
""
],
[
"Savini",
"Giorgio",
""
],
[
"Matsuo",
"Taro",
""
],
[
"Matsuo",
"Hiroshi",
""
],
[
"Spencer",
"Locke",
""
],
[
"Leisawitz",
"David T.",
""
]
] |
2512.10010
|
Ben-Chuan Lin
|
Shuo Wang, Ilaria Maccari, Xilin Feng, Ze-Nan Wu, Jia-Peng Peng, Kam Tuen Law, Y. X. Zhao, Andras Szabo, Andreas Schnyder, Ning Kang, Xiao-Song Wu, Jingchao Liu, Xuewen Fu, Mark H. Fischer, Manfred Sigrist, Dapeng Yu, Ben-Chuan Lin
|
Switchable half-quantum flux states in a ring of the kagome superconductor CsV$_3$Sb$_5$
|
49 pages, 23 figures
| null | null | null |
cond-mat.supr-con cond-mat.str-el
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Magnetic flux quantization in units of $\Phi_0 = h/2e$ is a defining feature of superconductivity, rooted in the charge-2e nature of Cooper pairs. In a ring geometry, the flux quantization leads to oscillations in the critical temperature with magnetic flux, known as the Little-Parks effect. While the maximal critical temperature is conventionally at zero flux, departures from this rule, for instance shifts by a half-quantum flux $\Phi_0/2$, clearly signal unconventional superconducting states and require sign-changing order parameters. Historically, such $\pi$-phase shifts in Little-Parks oscillations have been found in tricrystals or engineered ring structures that intentionally incorporate a $\pi$-phase shift. Here we report the discovery of switchable half-quantum flux states in rings made from single crystals of the kagome superconductor CsV$_3$Sb$_5$. We observe Little-Parks oscillations with a $\pi$-phase shift at zero bias current, which can be reversibly tuned to conventional Little-Parks oscillations upon applying a bias current. Between the $\pi$-phase and 0-phase regimes, $h/4e$ periodic oscillations appear. Our observations suggest unconventional pairing, potentially in the form of a multicomponent order parameter in the kagome superconductor CsV$_3$Sb$_5$, and reveal an electrically tunable landscape of competing superconducting condensates and fractional flux states.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:01:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wang",
"Shuo",
""
],
[
"Maccari",
"Ilaria",
""
],
[
"Feng",
"Xilin",
""
],
[
"Wu",
"Ze-Nan",
""
],
[
"Peng",
"Jia-Peng",
""
],
[
"Law",
"Kam Tuen",
""
],
[
"Zhao",
"Y. X.",
""
],
[
"Szabo",
"Andras",
""
],
[
"Schnyder",
"Andreas",
""
],
[
"Kang",
"Ning",
""
],
[
"Wu",
"Xiao-Song",
""
],
[
"Liu",
"Jingchao",
""
],
[
"Fu",
"Xuewen",
""
],
[
"Fischer",
"Mark H.",
""
],
[
"Sigrist",
"Manfred",
""
],
[
"Yu",
"Dapeng",
""
],
[
"Lin",
"Ben-Chuan",
""
]
] |
2512.10011
|
Lennart Landsmeer
|
Lennart P. L. Landsmeer, Amirreza Movahedin, Mario Negrello, Said Hamdioui, Christos Strydis
|
Spatial Spiking Neural Networks Enable Efficient and Robust Temporal Computation
| null | null | null | null |
cs.NE q-bio.NC
|
http://creativecommons.org/licenses/by/4.0/
|
The efficiency of modern machine intelligence depends on high accuracy with minimal computational cost. In spiking neural networks (SNNs), synaptic delays are crucial for encoding temporal structure, yet existing models treat them as fully trainable, unconstrained parameters, leading to large memory footprints, higher computational demand, and a departure from biological plausibility. In the brain, however, delays arise from physical distances between neurons embedded in space. Building on this principle, we introduce Spatial Spiking Neural Networks (SpSNNs), a framework in which neurons learn coordinates in a finite-dimensional Euclidean space and delays emerge from inter-neuron distances. This replaces per-synapse delay learning with position learning, substantially reducing parameter count while retaining temporal expressiveness. Across the Yin-Yang and Spiking Heidelberg Digits benchmarks, SpSNNs outperform SNNs with unconstrained delays despite using far fewer parameters. Performance consistently peaks in 2D and 3D networks rather than infinite-dimensional delay spaces, revealing a geometric regularization effect. Moreover, dynamically sparsified SpSNNs maintain full accuracy even at 90% sparsity, matching standard delay-trained SNNs while using up to 18x fewer parameters. Because learned spatial layouts map naturally onto hardware geometries, SpSNNs lend themselves to efficient neuromorphic implementation. Methodologically, SpSNNs compute exact delay gradients via automatic differentiation with custom-derived rules, supporting arbitrary neuron models and architectures. Altogether, SpSNNs provide a principled platform for exploring spatial structure in temporal computation and offer a hardware-friendly substrate for scalable, energy-efficient neuromorphic intelligence.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:01:39 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Landsmeer",
"Lennart P. L.",
""
],
[
"Movahedin",
"Amirreza",
""
],
[
"Negrello",
"Mario",
""
],
[
"Hamdioui",
"Said",
""
],
[
"Strydis",
"Christos",
""
]
] |
2512.10012
|
Mattes Mollenhauer
|
Mattes Mollenhauer, Christian Fiedler
|
Fuk-Nagaev inequality in smooth Banach spaces: Optimum bounds for distributions of heavy-tailed martingales
| null | null | null | null |
math.PR
|
http://creativecommons.org/licenses/by/4.0/
|
We derive a Fuk-Nagaev inequality for the maxima of norms of martingale sequences in smooth Banach spaces which allow for a finite number of higher conditional moments. The bound is obtained by combining an optimization approach for a Chernoff bound due to Rio (2017) with a classical bound for moment generating functions of smooth Banach space norms by Pinelis (1994). Our result improves comparable infinite-dimensional bounds in the literature by removing unnecessary centering terms and giving precise constants. As an application, we propose a McDiarmid-type bound for vector-valued functions which allow for a uniform bound on their conditional higher moments.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:02:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mollenhauer",
"Mattes",
""
],
[
"Fiedler",
"Christian",
""
]
] |
2512.10013
|
Mohammad Safdari
|
Mohammad Safdari
|
The distance to the boundary with respect to the Minkowski functional of a polytope
|
28 pages
| null | null | null |
math.MG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the regularity of the distance function to the boundary of a domain in $\mathbb{R}^n$, with respect to the Minkowski functional of a convex polytope. We obtain the regularity of the distance function in certain cases. We also explicitly compute the distance function in a collection of examples and observe the new interesting phenomena that arise for such distance functions.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:03:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Safdari",
"Mohammad",
""
]
] |
2512.10014
|
Yusuke Inoue
|
Yusuke Inoue, Keiichi Maeda, Takashi Nagao, Tatsuya Matsumoto
|
Formation of circumstellar material during double-white-dwarf mergers and the early excess emissions in Type Ia supernovae
|
14 pages, 8 figures, submitted to The Astrophysical Journal
| null | null | null |
astro-ph.HE astro-ph.SR
|
http://creativecommons.org/licenses/by/4.0/
|
Early excess emission observed in Type Ia supernovae (SNe Ia) within $\sim1$ day of explosion provides a critical window into their progenitor systems. In the present study, we investigate formation of the circumstellar matter (CSM) in double white-dwarf (WD) mergers. We further study the interaction between the CSM and the SN ejecta. We first model the orbital evolution and super-Eddington mass transfer/ejection in the double WD systems. We then conduct hydrodynamical and light-curve (LC) simulations of the SN-CSM interaction, assuming a prompt SN Ia explosion in a context of the carbon-ignited violent merger (C-ignited VM). Our simulations show that at the moment of the merger, the binary system has the CSM distribution following $\rho_{\mathrm{CSM}}\simeq D(r/10^{14}\ \mathrm{cm})^{-3.5}\ (D\simeq 10^{-14}\text{--}10^{-13}\ \rm g\ cm^{-3})$. The simulated LCs reproduce the early flux excesses across optical to UV bands, as well as their color evolution, observed in the VM candidates, i.e., 03fg/02es-like SNe Ia. This supports that 03fg/02es-like objects originate from the VM explosions. We also discuss the case of the helium-ignited VM, which might be realized in some WD-WD mergers depending on the He content in the system. Focused here is the timing when the explosion is initiated, and we find that the explosion is initiated after the companion WD is, at least partially, tidally disrupted also in this case; we thus expect the formation of the CSM through the mass transfer phase also for the helium-ignited VM scenario.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:06:46 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Inoue",
"Yusuke",
""
],
[
"Maeda",
"Keiichi",
""
],
[
"Nagao",
"Takashi",
""
],
[
"Matsumoto",
"Tatsuya",
""
]
] |
2512.10015
|
Richard Ruiz
|
Trina Basu and Richard Ruiz
|
The Four Polarizations of the $W$ at High Energies
|
60 pages (incl. appendix and refs), 13 figures (36 image files), comments welcomed!
| null | null |
IFJPAN-IV-2025-23, COMETA-2025-45
|
hep-ph hep-ex hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate polarization-induced interference and off-shell effects in predictions for high-energy, multi-leg processes with intermediate weak bosons carrying fixed helicities. Building on the ``truncated propagator'' paradigm, we carry out our analysis at the level of helicity amplitudes and squared amplitudes. We introduce bookkeeping devices, suitable for covariant and axial gauge choices, that simplify the analytical evaluation of polarized amplitudes, and make power counting of mass-over-energy factors more manifest. Among other results, we show that polarization interference (i) is generally non-zero, even in on-shell limits, (ii) can be negative and comparable to longitudinal contributions, and (iii) is generated by helicity inversion and therefore suppressed (or zero) in high-energy limits for $s$- and $t$-channel exchanges. Connections between gauge invariance and the scalar polarization are also discussed, as is a scheme for reducing gauge dependence in predictions for polarized scattering rates. As case studies, we consider charged-current processes, including $W$(+jets), top quark decay, and neutrino deep-inelastic scattering.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:07:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Basu",
"Trina",
""
],
[
"Ruiz",
"Richard",
""
]
] |
2512.10016
|
Marvin Alles
|
Marvin Alles, Xingyuan Zhang, Patrick van der Smagt, Philip Becker-Ehmck
|
Latent Action World Models for Control with Unlabeled Trajectories
| null | null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Inspired by how humans combine direct interaction with action-free experience (e.g., videos), we study world models that learn from heterogeneous data. Standard world models typically rely on action-conditioned trajectories, which limits effectiveness when action labels are scarce. We introduce a family of latent-action world models that jointly use action-conditioned and action-free data by learning a shared latent action representation. This latent space aligns observed control signals with actions inferred from passive observations, enabling a single dynamics model to train on large-scale unlabeled trajectories while requiring only a small set of action-labeled ones. We use the latent-action world model to learn a latent-action policy through offline reinforcement learning (RL), thereby bridging two traditionally separate domains: offline RL, which typically relies on action-conditioned data, and action-free training, which is rarely used with subsequent RL. On the DeepMind Control Suite, our approach achieves strong performance while using about an order of magnitude fewer action-labeled samples than purely action-conditioned baselines. These results show that latent actions enable training on both passive and interactive data, which makes world models learn more efficiently.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:09:45 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Alles",
"Marvin",
""
],
[
"Zhang",
"Xingyuan",
""
],
[
"van der Smagt",
"Patrick",
""
],
[
"Becker-Ehmck",
"Philip",
""
]
] |
2512.10017
|
Jeffrey Shallit
|
Delaram Moradi, Narad Rampersad, and Jeffrey Shallit
|
Complexity of Linear Subsequences of $k$-Automatic Sequences
| null | null | null | null |
cs.FL cs.DM math.CO math.NT
|
http://creativecommons.org/licenses/by/4.0/
|
We construct automata with input(s) in base $k$ recognizing some basic relations and study their number of states. We also consider some basic operations on $k$-automatic sequences and discuss their state complexity. We find a relationship between subword complexity of the interior sequence $(h'(i))_{i \geq 0}$ and state complexity of the linear subsequence $(h(ni+c))_{i \geq 0}$. We resolve a recent question of Zantema and Bosma about linear subsequences of $k$-automatic sequences with input in most-significant-digit-first format. We also discuss the state complexity and runtime complexity of using a reasonable interpretation of B\"uchi arithmetic to actually construct some of the studied automata recognizing relations or carrying out operations on automatic sequences.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:10:26 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Moradi",
"Delaram",
""
],
[
"Rampersad",
"Narad",
""
],
[
"Shallit",
"Jeffrey",
""
]
] |
2512.10018
|
Magnus Goffeng
|
Heiko Gimperlein, Magnus Goffeng
|
Sharp mapping properties of Poisson transforms and the Baum-Connes conjecture
|
41 pages
| null | null | null |
math.KT math.AP math.OA math.RT
|
http://creativecommons.org/licenses/by/4.0/
|
We prove a sharp, quantitative analogue of Helgason's conjecture at the level of distributions: For a semisimple Lie group $G$ of real rank one, Poisson transforms map a Sobolev space on $P\backslash G$ boundedly with closed range to an $L^2$-space on $K\backslash G$. The result is obtained for the Poisson transform studied by Knapp-Wallach under the name Szeg\"o map, and the appropriate Sobolev spaces are defined using van Erp-Yuncken's Heisenberg calculus. The proof generalizes to show that commutators of this Poisson transform with smooth functions on the Furstenberg compactification are compact. This proves the remaining open conjecture in Julg's seminal program to establish the Baum-Connes conjecture for closed subgroups of semisimple Lie groups of real rank one.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:11:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gimperlein",
"Heiko",
""
],
[
"Goffeng",
"Magnus",
""
]
] |
2512.10019
|
Aarya Patil
|
Aarya A. Patil, Conny Aerts, Nikki Y. N. Wang, Jordan Van Beeck, and May G. Pedersen
|
Beyond prewhitening: detection of gravity modes and their period spacings in slowly pulsating B stars using the multitaper F-test
|
14 pages, 10 figures, 2 tables (plus 11 pages of figures in Appendix), Submitted to A&A
| null | null | null |
astro-ph.SR astro-ph.IM stat.AP
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Gravity modes in main-sequence stars have traditionally been studied using a prewhitening approach, which iteratively identifies modes in the Fourier domain and subsequently tunes their frequencies, amplitudes, and phases through time-domain regression. While effective, this method becomes inefficient when analysing large volumes of long time-series data and often relies on subjective stopping criteria to determine the number of iterations. We aim to perform frequency extraction of gravity modes in slowly pulsating B (SPB) stars using a statistically robust, data-driven approach based on advanced power spectrum and harmonic analysis techniques. Our approach employs the multitaper non-uniform fast Fourier transform, mtNUFFT, a power spectrum estimator that addresses several statistical limitations of traditional methods such as the Lomb-Scargle periodogram. We apply its extension, the multitaper F-test, to extract coherent gravity modes from 4-year Kepler light curves of SPB stars and to search for period spacing patterns among the extracted modes. The multitaper F-test enables fast and accurate extraction of the properties of gravity modes with quasi-infinite lifetimes, preferentially selecting modes that exhibit purely periodic behaviour. Although the method typically extracts fewer frequencies than conventional prewhitening, it recovers most known modes and, in some cases, reveals new ones. We also find evidence for gravity modes with long but finite lifetimes, and detect more than one period spacing pattern in some of the studied SPB stars. Overall, the multitaper F-test offers a more objective and statistically sound alternative to prewhitening. It scales efficiently to large datasets containing thousands of pulsators, and has the potential to facilitate mode identification and to distinguish between the different excitation mechanisms operating in SPB stars.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:17:23 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Patil",
"Aarya A.",
""
],
[
"Aerts",
"Conny",
""
],
[
"Wang",
"Nikki Y. N.",
""
],
[
"Van Beeck",
"Jordan",
""
],
[
"Pedersen",
"May G.",
""
]
] |
2512.10020
|
Nitin Awathare
|
Ayush Nainwal, Atharva Kamble, Nitin Awathare
|
A Comparative Analysis of zk-SNARKs and zk-STARKs: Theory and Practice
| null | null | null | null |
cs.CR cs.DC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Zero-knowledge proofs (ZKPs) are central to secure and privacy-preserving computation, with zk-SNARKs and zk-STARKs emerging as leading frameworks offering distinct trade-offs in efficiency, scalability, and trust assumptions. While their theoretical foundations are well studied, practical performance under real-world conditions remains less understood.
In this work, we present a systematic, implementation-level comparison of zk-SNARKs (Groth16) and zk-STARKs using publicly available reference implementations on a consumer-grade ARM platform. Our empirical evaluation covers proof generation time, verification latency, proof size, and CPU profiling. Results show that zk-SNARKs generate proofs 68x faster with 123x smaller proof size, but verify slower and require trusted setup, whereas zk-STARKs, despite larger proofs and slower generation, verify faster and remain transparent and post-quantum secure. Profiling further identifies distinct computational bottlenecks across the two systems, underscoring how execution models and implementation details significantly affect real-world performance. These findings provide actionable insights for developers, protocol designers, and researchers in selecting and optimizing proof systems for applications such as privacy-preserving transactions, verifiable computation, and scalable rollups.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:19:47 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Nainwal",
"Ayush",
""
],
[
"Kamble",
"Atharva",
""
],
[
"Awathare",
"Nitin",
""
]
] |
2512.10021
|
Sizhe Cheng
|
Sizhe Cheng, Devadyouti Das, Mykhaylo Barchuk, Raveen Armstrong, Michele M. Klingbeil, Becca Thomases, Shuang Zhou
|
Corkscrew motion of Trypanosome brucei is driven by helical beating of the flagellum and facilitated by its bent shape
|
41 pages, 5 figures in manuscript, 10 figures in supplementary materials
| null | null | null |
physics.bio-ph physics.flu-dyn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In the pathogenic parasite Trypanosoma brucei, a laterally attached flagellum drives rapid deformation of the complex cell body, producing puzzling dynamics. High-speed defocusing imaging reveals that surface points trace flower-like patterns in transverse planes. The petals arise from clockwise flagellar beating, which generates a right-handed helical wave propagating from the anterior tip along the body, advancing the cell like a twisted corkscrew. The central lobes result from slower counterclockwise body rotation required to balance the active torque. The bent cell shape underneath the flagellum superimposes these two chiral motions at different radial distances, producing the observed patterns. Three-dimensional hydrodynamic simulations using the method of regularized Stokeslets reproduce these dynamics and show that bent cell shape enhances swimming, suggesting an adaptive advantage of T. brucei's morphology.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:21:38 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Cheng",
"Sizhe",
""
],
[
"Das",
"Devadyouti",
""
],
[
"Barchuk",
"Mykhaylo",
""
],
[
"Armstrong",
"Raveen",
""
],
[
"Klingbeil",
"Michele M.",
""
],
[
"Thomases",
"Becca",
""
],
[
"Zhou",
"Shuang",
""
]
] |
2512.10022
|
Jim Geelen
|
Jim Geelen
|
Taking minors without splitting tangles
| null | null | null | null |
math.CO
|
http://creativecommons.org/licenses/by/4.0/
|
We prove that any element in a matroid can be removed, by either deletion or contraction, in such a way that no tangle "splits".
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:23:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Geelen",
"Jim",
""
]
] |
2512.10023
|
Kyle Neumann
|
Kyle D. Neumann, Abraham D. Falcone, Stephen DiKerby, Sierra Deppe, Elizabeth C. Ferrara, Jamie A. Kennea, Brad Cenko, and Eric Grove
|
Classification of a New X-ray Catalog of Likely Counterparts to 4FGL-DR4 Unassociated Gamma-ray Sources Using a Neural Network
|
19 Pages, 8 Figures, 9 Tables. Data presented in this submission is included in machine-readable format as 2 ancillary files. Accepted for publication in ApJ
| null | null | null |
astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
Our survey of the fourth $\mathit{Fermi}$ Large Area Telescope catalog (4FGL) unassociated gamma-ray source regions using the X-Ray Telescope (XRT) and Ultraviolet/Optical Telescope (UVOT) aboard the Neil Gehrels $\mathit{Swift}$ Observatory ($\mathit{Swift}$) provides new XRT and UVOT source detections and localizations to help identify potential low-energy counterparts to unassociated $\mathit{Fermi}$ gamma-ray sources. We present a catalog of 218 singlet and 70 multiplet $\mathit{Swift}$ X-ray sources detected within the positional uncertainty ellipses of 244 unassociated $\mathit{Fermi}$ gamma-ray sources from the 4FGL-DR4 catalog, 144 of which are not previously cataloged by Kerby et al. (2021b). For each X-ray source, we derive its X-ray flux and photon index, then use simultaneous UVOT observations with optical survey data to estimate its $V$-band magnitude. We use these parameters as inputs for a multi-layer perceptron (MLP) neural network classifier (NNC) trained to classify sources as blazars, pulsars, or ambiguous gamma-ray sources. For the 213 singlet sources with X-ray and optical data, we classify 173 as likely blazars ($P_\mathrm{bzr} > 0.99$) and 6 as likely pulsars ($P_\mathrm{bzr} < 0.01$), with 34 sources yielding ambiguous results. Including 70 multiplet X-ray sources, we increase the number of $P_\mathrm{bzr} > 0.99$ to 227 and $P_\mathrm{bzr} < 0.01$ to 16. For the subset of these classifications that have been previously studied, a large majority agree with prior classifications, supporting the validity of using this NNC to classify the unknown and newly detected gamma-ray sources.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:24:13 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Neumann",
"Kyle D.",
""
],
[
"Falcone",
"Abraham D.",
""
],
[
"DiKerby",
"Stephen",
""
],
[
"Deppe",
"Sierra",
""
],
[
"Ferrara",
"Elizabeth C.",
""
],
[
"Kennea",
"Jamie A.",
""
],
[
"Cenko",
"Brad",
""
],
[
"Grove",
"Eric",
""
]
] |
2512.10024
|
Anna Frid
|
Anna E. Frid
|
Small palindromic lengths in free groups and word equations with antimorphisms
| null | null | null | null |
math.CO math.GR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The palindromic length of a finite word $w$ is defined as the minimal number of palindromes such that their product is $w$. Clearly, this function may take different values depending on if we consider $w$ as an element a free semigroup or of a free group: for example, in the free semigroup, the palindromic length of $abca$ is 4 (here every letter is a palindrome), and in the free group, it is 3 since $abca=(aba)(a^{-1}a^{-1})(aca)$.
In free semigroups, the palindromic length can clearly be computed, and there are fast algorithms for that. In free groups, the question is trickier. In this paper, we characterize words in the free group whose palindromic length is 2 and 3.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:25:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Frid",
"Anna E.",
""
]
] |
2512.10025
|
Georgios Tsikalas
|
Nikolaos Chalmoukis, Georgios Tsikalas, Dmitry Yakubovich
|
Operators with small Kreiss constants
| null | null | null | null |
math.FA math.CV math.SP
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate matrices satisfying the Kreiss condition
$$\|(zI-T)^{-1}\|\le\cfrac{K}{|z|-1}, \hspace{0.7 cm} |z|>1, $$
with $K$ lying arbitrarily close to $1.$ We provide lower bounds for the power growth of such matrices, which complement and refine related estimates due to Nikolski and Spijker-Tracogna-Welfert. We also study operators that satisfy a variant of the above Kreiss condition where $K$ is replaced by $1+\epsilon(|z|)$, where the positive continuous function $\epsilon(|z|)$ tends to $0$ as $|z|\to 1^+.$ We show that, if the spectum of $T$ touches the unit circle only at a single point and the resolvent of $T$ satisfies a growth restriction along the unit circle, it is possible to choose $\epsilon$ so that this Kreiss-type condition guarantees similarity to a contraction. At the core of our proof lies a positivity argument involving the double-layer potential operator. Counterexamples related to less restrictive choices of $\epsilon$ are also provided.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:27:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chalmoukis",
"Nikolaos",
""
],
[
"Tsikalas",
"Georgios",
""
],
[
"Yakubovich",
"Dmitry",
""
]
] |
2512.10026
|
Carlota Andres
|
Carlota Andres, Jack Holguin, Benjamin Kimelman, Raghav Kunnawalkam Elayavalli, Jussi Viinikainen, Zhong Yang
|
Living on the edge: radius effects in the angular substructure of heavy-ion jets
|
14 pages, 12 edgy figures, plus 2 appendices with 3 more equally edgy figures
| null | null |
MIT-CTP/5982
|
hep-ph hep-ex nucl-ex nucl-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Jet substructure observables serve as essential tools for probing the quark-gluon plasma produced in relativistic heavy-ion collisions. Their interpretation, however, is often complicated by edge effects, which arise when correlated particles fall outside the reconstructed jet radius, introducing distortions that obscure the underlying QCD dynamics. In this work, we present a comprehensive phenomenological study of edge effects in soft-insensitive angular observables, taking the two-point energy correlator (EEC) as a representative example. We argue that these distortions scale linearly with the average angular separation between the winner-take-all and $E$-scheme axes $\langle \phi \rangle$, and validate this behavior across proton-proton (p-p) simulations with Pythia8 and Herwig7, as well as lead-lead (Pb-Pb) simulations using JEWEL and CoLBT. In p-p collisions, edge effects are strongly suppressed, scaling as $(R_L/R)^4$, whereas medium-modified jets can exhibit larger distortions, with contributions scaling as $(R_L/R)^2$ and $(R_L/R)^4$. Taking Pb-Pb/p-p ratios of the EEC substantially reduces, but does not completely eliminate, these distortions, highlighting the need of accounting for edge effects in the interpretation of heavy-ion jet substructure measurements. Since edge effects are largely governed by the $\langle \phi \rangle$ distribution, studying this distribution provides a new handle for benchmarking and constraining the modeling of edge effects in heavy-ion event generators.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:31:00 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Andres",
"Carlota",
""
],
[
"Holguin",
"Jack",
""
],
[
"Kimelman",
"Benjamin",
""
],
[
"Elayavalli",
"Raghav Kunnawalkam",
""
],
[
"Viinikainen",
"Jussi",
""
],
[
"Yang",
"Zhong",
""
]
] |
2512.10027
|
Yassin Bahid
|
Yassin Bahid, Eduardo Corona, Nancy Rodriguez
|
A Mass Preserving Numerical Scheme for Kinetic Equations that Model Social Phenomena
| null | null | null | null |
math.NA cs.NA
|
http://creativecommons.org/licenses/by/4.0/
|
In recent years, kinetic equations have been used to model many social phenomena. A key feature of these models is that transition rate kernels involve Dirac delta functions, which capture sudden, discontinuous state changes. Here, we study kinetic equations with transition rates of the form $$ T(x,y,u) = \delta_{\phi(x,y) - u}. $$ We establish the global existence and uniqueness of solutions for these systems and introduce a fully deterministic scheme, the \emph{Mass Preserving Collocation Method}, which enables efficient, high fidelity simulation of models with multiple subsystems. We validate the accuracy, efficiency, and consistency of the solver on models with up to five subsystems, and compare its performance against two state-of-the-art agent-based methods: Tau-leaping and hybrid methods. Our scheme resolves subsystem distributions captured by these stochastic approaches while preserving mass numerically, requiring significantly less computational time and resources, and avoiding variability and hyperparameter tuning characteristic of these methods.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:31:25 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bahid",
"Yassin",
""
],
[
"Corona",
"Eduardo",
""
],
[
"Rodriguez",
"Nancy",
""
]
] |
2512.10028
|
Yazid Delenda
|
Achwaq Ghezal, Yazid Delenda, Mekki Aouachria
|
Constraining non-commutative geometry with W/Z+jet production at the LHC
|
21 pages, 8 figures
| null | null | null |
hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We present a comprehensive calculation of the squared matrix elements for all partonic channels contributing to $W^\pm/Z$+jet production at hadron colliders within the framework of the non-commutative Standard Model (NCSM), including leptonic decays $W\to e\nu$ and $Z\to e^+e^-$. Our computation incorporates both $\mathcal{O}(\Theta)$ corrections to the Standard Model vertices and additional interaction terms inherent to the NCSM. A key finding is that the production amplitudes receive first-order corrections at $\mathcal{O}(\Theta)$, a distinctive feature compared to many other processes where non-commutative effects enter only at $\mathcal{O}(\Theta^2)$. The leptonic decay widths, in contrast, are modified solely at $\mathcal{O}(\Theta^2)$. This $\mathcal{O}(\Theta)$ enhancement provides improved sensitivity to non-commutative geometry, allowing us to probe for and constrain the non-commutative energy scale in the multi-TeV range. We provide numerical predictions for angular (azimuthal and rapidity) distributions and the forward--backward asymmetry, and compare them to state-of-the-art Standard Model predictions at leading and next-to-leading order from the \texttt{MCFM} Monte Carlo program. Finally, we test the NCSM with experimental data by analyzing an unbinned, particle-level $Z$+jet dataset from the ATLAS experiment. From this data, we calculate the azimuthal spectrum and forward-backward asymmetry, which are then used to derive stringent lower bounds on the non-commutative scale $\Lambda$.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:32:24 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ghezal",
"Achwaq",
""
],
[
"Delenda",
"Yazid",
""
],
[
"Aouachria",
"Mekki",
""
]
] |
2512.10029
|
Shresta Bellary Seetharam
|
Shresta B.Seetharam, Mohamed Nabeel, William Melicher
|
Malicious GenAI Chrome Extensions: Unpacking Data Exfiltration and Malicious Behaviours
|
VIRUS BULLETIN CONFERENCE SEPTEMBER 2025
| null | null | null |
cs.CR
|
http://creativecommons.org/licenses/by/4.0/
|
The rapid proliferation of AI and GenAI tools has extended to the Chrome Web Store. Cybercriminals are exploiting this trend, deploying malicious Chrome extensions posing as AI tools or impersonating popular GenAI models to target users. These extensions often appear legitimate while secretly exfiltrating sensitive data or redirecting users web traffic to attacker-controlled domains.
To examine the impact of this trend on the browser extension ecosystem, we curated a dataset of 5,551 AI-themed extensions released over a nine-month period to the Chrome Web Store. Using a multi-signal detection methodology that combines manifest analysis, domain reputation, and runtime network behavior, supplemented with human review, we identified 154 previously undetected malicious Chrome extensions. Together with extensions known from public threat research disclosures, this resulted in a final set of 341 malicious extensions for analysis. Of these, 29 were GenAI-related, forming the focus of our in-depth analysis and disclosure.
We deconstruct representative GenAI cases, including Supersonic AI, DeepSeek AI | Free AI Assistant, and Perplexity Search, to illustrate attacker techniques such as Adversary-in-the-Browser, impersonation, bait-and-switch updates, query hijacking, and redirection. Our findings show that threat actors are leveraging GenAI trends and exploiting browser extension APIs and settings for malicious purposes. This demonstrates that the browser extension threat landscape is directly evolving alongside the rapid adoption of GenAI technologies.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:33:58 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Seetharam",
"Shresta B.",
""
],
[
"Nabeel",
"Mohamed",
""
],
[
"Melicher",
"William",
""
]
] |
2512.10030
|
Pavel Kuptsov
|
Pavel V. Kuptsov
|
An improved approach for estimating the dimension of inertial manifolds in chaotic distributed dynamical systems via analysis of angles between tangent subspaces
|
11 pages, 13 figures
| null | null | null |
nlin.CD math.DS
|
http://creativecommons.org/licenses/by/4.0/
|
While a previously proposed method for estimating inertial manifold dimension, based on explicitly computing angles between pairs of covariant Lyapunov vectors (CLVs), employs efficient algorithms, it remains computationally demanding due to its substantial resource requirements. In this work, we introduce an improved method to determine this dimension by analyzing the angles between tangent subspaces spanned by the CLVs. This approach builds upon a fast numerical technique for assessing chaotic dynamics hyperbolicity. Crucially, the proposed method requires significantly less computational effort and minimizes memory usage by eliminating the need for explicit CLV computation. We test our method on two canonical systems: the complex Ginzburg-Landau equation and a diffusively coupled chain of Lorenz oscillators. For the former, the results confirm the accuracy of the new approach by matching prior dimension estimates. For the latter, the analysis demonstrates the absence of a low-dimensional inertial manifold, highlighting a complex regime that merits further investigation. The presented method offers a practical and efficient tool for characterizing high-dimensional attractors in extended dynamical systems.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:35:30 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kuptsov",
"Pavel V.",
""
]
] |
2512.10031
|
Woojin Lee
|
Woojin Lee, Hyugjae Chang, Jaeho Moon, Jaehyup Lee, Munchurl Kim
|
ABBSPO: Adaptive Bounding Box Scaling and Symmetric Prior based Orientation Prediction for Detecting Aerial Image Objects
|
17 pages, 11 figures, 8 tables, supplementary included. Accepted to CVPR 2025. Please visit our project page at https://kaist-viclab.github.io/ABBSPO_site/
| null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Weakly supervised oriented object detection (WS-OOD) has gained attention as a cost-effective alternative to fully supervised methods, providing both efficiency and high accuracy. Among weakly supervised approaches, horizontal bounding box (HBox)-supervised OOD stands out for its ability to directly leverage existing HBox annotations while achieving the highest accuracy under weak supervision settings. This paper introduces adaptive bounding box scaling and symmetry-prior-based orientation prediction, called ABBSPO, a framework for WS-OOD. Our ABBSPO addresses limitations of previous HBox-supervised OOD methods, which compare ground truth (GT) HBoxes directly with the minimum circumscribed rectangles of predicted RBoxes, often leading to inaccurate scale estimation. To overcome this, we propose: (i) Adaptive Bounding Box Scaling (ABBS), which appropriately scales GT HBoxes to optimize for the size of each predicted RBox, ensuring more accurate scale prediction; and (ii) a Symmetric Prior Angle (SPA) loss that exploits inherent symmetry of aerial objects for self-supervised learning, resolving issues in previous methods where learning collapses when predictions for all three augmented views (original, rotated, and flipped) are consistently incorrect. Extensive experimental results demonstrate that ABBSPO achieves state-of-the-art performance, outperforming existing methods.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:37:54 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Lee",
"Woojin",
""
],
[
"Chang",
"Hyugjae",
""
],
[
"Moon",
"Jaeho",
""
],
[
"Lee",
"Jaehyup",
""
],
[
"Kim",
"Munchurl",
""
]
] |
2512.10032
|
Jan Marco Ruiz De Vargas
|
Jan Marco Ruiz de Vargas, Kirtan Padh, Niki Kilbertus
|
Cluster-Dags as Powerful Background Knowledge For Causal Discovery
|
23 pages, 5 figures
| null | null | null |
cs.LG cs.AI stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Finding cause-effect relationships is of key importance in science. Causal discovery aims to recover a graph from data that succinctly describes these cause-effect relationships. However, current methods face several challenges, especially when dealing with high-dimensional data and complex dependencies. Incorporating prior knowledge about the system can aid causal discovery. In this work, we leverage Cluster-DAGs as a prior knowledge framework to warm-start causal discovery. We show that Cluster-DAGs offer greater flexibility than existing approaches based on tiered background knowledge and introduce two modified constraint-based algorithms, Cluster-PC and Cluster-FCI, for causal discovery in the fully and partially observed setting, respectively. Empirical evaluation on simulated data demonstrates that Cluster-PC and Cluster-FCI outperform their respective baselines without prior knowledge.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:39:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"de Vargas",
"Jan Marco Ruiz",
""
],
[
"Padh",
"Kirtan",
""
],
[
"Kilbertus",
"Niki",
""
]
] |
2512.10033
|
Sarwan Ali
|
Sarwan Ali
|
Robust Gradient Descent via Heavy-Ball Momentum with Predictive Extrapolation
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Accelerated gradient methods like Nesterov's Accelerated Gradient (NAG) achieve faster convergence on well-conditioned problems but often diverge on ill-conditioned or non-convex landscapes due to aggressive momentum accumulation. We propose Heavy-Ball Synthetic Gradient Extrapolation (HB-SGE), a robust first-order method that combines heavy-ball momentum with predictive gradient extrapolation. Unlike classical momentum methods that accumulate historical gradients, HB-SGE estimates future gradient directions using local Taylor approximations, providing adaptive acceleration while maintaining stability. We prove convergence guarantees for strongly convex functions and demonstrate empirically that HB-SGE prevents divergence on problems where NAG and standard momentum fail. On ill-conditioned quadratics (condition number $\kappa=50$), HB-SGE converges in 119 iterations while both SGD and NAG diverge. On the non-convex Rosenbrock function, HB-SGE achieves convergence in 2,718 iterations where classical momentum methods diverge within 10 steps. While NAG remains faster on well-conditioned problems, HB-SGE provides a robust alternative with speedup over SGD across diverse landscapes, requiring only $O(d)$ memory overhead and the same hyperparameters as standard momentum.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:39:43 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ali",
"Sarwan",
""
]
] |
2512.10034
|
Cassandra Masschelein
|
Salom\'e Guilbert, Cassandra Masschelein, Jeremy Goumaz, Bohdan Naida, Philippe Schwaller
|
DynaMate: An Autonomous Agent for Protein-Ligand Molecular Dynamics Simulations
| null | null | null | null |
cs.AI cs.CE
|
http://creativecommons.org/licenses/by/4.0/
|
Force field-based molecular dynamics (MD) simulations are indispensable for probing the structure, dynamics, and functions of biomolecular systems, including proteins and protein-ligand complexes. Despite their broad utility in drug discovery and protein engineering, the technical complexity of MD setup, encompassing parameterization, input preparation, and software configuration, remains a major barrier for widespread and efficient usage. Agentic LLMs have demonstrated their capacity to autonomously execute multi-step scientific processes, and to date, they have not successfully been used to automate protein-ligand MD workflows. Here, we present DynaMate, a modular multi-agent framework that autonomously designs and executes complete MD workflows for both protein and protein-ligand systems, and offers free energy binding affinity calculations with the MM/PB(GB)SA method. The framework integrates dynamic tool use, web search, PaperQA, and a self-correcting behavior. DynaMate comprises three specialized modules, interacting to plan the experiment, perform the simulation, and analyze the results. We evaluated its performance across twelve benchmark systems of varying complexity, assessing success rate, efficiency, and adaptability. DynaMate reliably performed full MD simulations, corrected runtime errors through iterative reasoning, and produced meaningful analyses of protein-ligand interactions. This automated framework paves the way toward standardized, scalable, and time-efficient molecular modeling pipelines for future biomolecular and drug design applications.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:40:51 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Guilbert",
"Salomé",
""
],
[
"Masschelein",
"Cassandra",
""
],
[
"Goumaz",
"Jeremy",
""
],
[
"Naida",
"Bohdan",
""
],
[
"Schwaller",
"Philippe",
""
]
] |
2512.10035
|
U\u{g}ur Te\u{g}in
|
Firdevs Y\"uce, Bora \c{C}arp{\i}nl{\i}o\u{g}lu, U\u{g}ur Te\u{g}in
|
The role of modes in nonlinear fiber optical computing
|
5 pages, 4 figures
| null | null | null |
physics.optics
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We investigate the nonlinear propagation of light in graded-index multimode fiber, utilizing it as an optical computing unit, and quantify how it employs waveguide modes to process information. Using a time-dependent spatiotemporal propagation model with modal decomposition, we evaluate several benchmark regression and classification tasks and study the modal content of the generated speckles, which couples with a simple digital layer to perform optical computing. Analysis of modal entropy and energy-based mode counts reveals that effective computation is confined to a low-dimensional modal subspace, whose identity depends on the task and propagation regime. This also sets a trade-off between modal richness and nonlinear beam self-cleaning. These results establish modal statistics as practical design metrics for fiber-based optical computers.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:42:11 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yüce",
"Firdevs",
""
],
[
"Çarpınlıoğlu",
"Bora",
""
],
[
"Teğin",
"Uğur",
""
]
] |
2512.10036
|
Nimesha Gunasekara
|
Nimesha Gunasekara and Ebrahim Bedeer
|
Analysis and Compensation of Receiver IQ Imbalance and Residual CFO Error for AFDM
| null | null | null | null |
eess.SP
|
http://creativecommons.org/licenses/by/4.0/
|
Affine frequency division multiplexing (AFDM) is a promising waveform for future wireless communication systems. In this paper, we analyze the impact of receiver in-phase and quadrature (IQ) imbalance and residual carrier frequency offset (CFO) error on AFDM signals. Our analysis shows that the receiver IQ imbalance may not preserve the sparsity of the AFDM effective channel matrix because of the complex-conjugate operator of the discrete affine Fourier transform (DAFT). Moreover, the residual CFO error causes energy leakage in the effective channel matrix in the affine domain. To mitigate these effects, we extend the linear minimum mean-square error (LMMSE) detector to handle the improper Gaussian noise arising from the receiver IQ imbalance. Simulation results demonstrate that the proposed LMMSE detector effectively compensates for the receiver hardware impairments.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:42:35 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gunasekara",
"Nimesha",
""
],
[
"Bedeer",
"Ebrahim",
""
]
] |
2512.10037
|
Pedro Henrique Rocha De Andrade
|
Pedro Henrique Rocha de Andrade, Maria Luiza Linhares Dantas and Ana Cecilia Soja
|
Stellar properties and chemical features of the Gaia Catalogue of Nearby Stars observed by GALAH DR4
|
To appear in the proceedings of the Brazilian Astronomical Society
| null | null | null |
astro-ph.SR astro-ph.GA
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
The Gaia Catalogue of Nearby Stars (GCNS) comprises approximately 330 000 stars within 100 pc of the Sun, as observed by Gaia data release 3 (Gaia DR3). Meanwhile, the GALAH DR4 survey has spectroscopically characterised nearly one million stars, delivering detailed chemical abundances (up to 30 elements). We present a joint analysis of the $\sim$ 6 000 stars common to both catalogues, offering initial insights into the stellar and chemical properties of the solar neighborhood. Our preliminary results indicate that the majority of these stars are FGK main-sequence objects, with some A-type interlopers (with effective temperatures ranging between 3 000 and 8 000 K), with median ages of $\sim$ 1.6 Gyr (ranging from 0.10 to 14.79 Gyr), and metal-poorer when compared to the Sun: [Fe/H] $\approx$ -0.19 dex. Additionally, most of the stars are disc members, with some local halo (high-velocity) stars identified. Building on this foundation, future work will deeper exploit the full spectroscopic information and orbital parameters from value-added catalogues to refine Galactic component classifications (thin-thick disc versus halo membership), perform detailed chemical profiling, and deliver a comprehensive chemo-dynamical characterisation of the solar neighborhood. This will provide new insights into the formation and evolution of nearby stellar populations.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:42:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"de Andrade",
"Pedro Henrique Rocha",
""
],
[
"Dantas",
"Maria Luiza Linhares",
""
],
[
"Soja",
"Ana Cecilia",
""
]
] |
2512.10038
|
Jia Cheng Hu
|
Jia Cheng Hu, Roberto Cavicchioli, Alessandro Capotondi
|
Diffusion Is Your Friend in Show, Suggest and Tell
| null |
2025 IEEE International Conference on Big Data
| null | null |
cs.CV cs.CL
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Diffusion Denoising models demonstrated impressive results across generative Computer Vision tasks, but they still fail to outperform standard autoregressive solutions in the discrete domain, and only match them at best. In this work, we propose a different paradigm by adopting diffusion models to provide suggestions to the autoregressive generation rather than replacing them. By doing so, we combine the bidirectional and refining capabilities of the former with the strong linguistic structure provided by the latter. To showcase its effectiveness, we present Show, Suggest and Tell (SST), which achieves State-of-the-Art results on COCO, among models in a similar setting. In particular, SST achieves 125.1 CIDEr-D on the COCO dataset without Reinforcement Learning, outperforming both autoregressive and diffusion model State-of-the-Art results by 1.5 and 2.5 points. On top of the strong results, we performed extensive experiments to validate the proposal and analyze the impact of the suggestion module. Results demonstrate a positive correlation between suggestion and caption quality, overall indicating a currently underexplored but promising research direction. Code will be available at: https://github.com/jchenghu/show\_suggest\_tell.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:44:19 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Hu",
"Jia Cheng",
""
],
[
"Cavicchioli",
"Roberto",
""
],
[
"Capotondi",
"Alessandro",
""
]
] |
2512.10039
|
Cristian Vay
|
Jack Arce, Cristian Vay
|
Lifting in a non semisimple world
|
17 pages
| null | null | null |
math.QA
|
http://creativecommons.org/licenses/by/4.0/
|
This is a contribution to the problem of classifying all deformations - a. k. a. liftings - of the bosonization of a Nichols algebra $\mathfrak{B}(V)$ over a cosemisimple and non-semisimple Hopf algebra $H$. Such a situation arises when the underlying field has positive characteristic or when $H$ is infinite-dimensional. Given an $H$-module $M$ that is an extension of $V$ by $H$, we first introduce an algebra $T(V)\#_MH$ which generalizes the usual bosonization $T(V)\#H$. Indeed, these two objects coincide when $M$ is a trivial extension. We provide necessary conditions for $T(V)\#_MH$ to be a Hopf algebra and a cocycle deformation of $T(V)\#H$. These conditions appear particularly natural when $H$ is a group algebra. We then prove that every lifting is a quotient of $T(V)\#_MH$ for some extension $M$. Echoing Archimedes, $T(V)\#_MH$ stands as a fulcrum over which we can pivot to lift the relations of the Nichols algebra. From this point, one can replicate the strategy proposed by Andruskiewitsch, Angiono, Garcia I., Masuoka and the second author to show that every lifting is a cocycle deformation of $\mathfrak{B}(V)\#H$. We illustrate this idea with two examples of different nature. We classify all pointed liftings of the Fomin-Kirillov algebra $\mathcal{FK}_3$ in characteristic $2$, and prove they are all cocycle deformations one another. We also prove that the Jordanian enveloping algebra of $\mathfrak{sl}(2)$ defined by Andruskiewitsch, Angiono and Heckenberger is a cocycle deformation of the bosonization of the Jordan plane over the infinite cyclic group.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:45:06 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Arce",
"Jack",
""
],
[
"Vay",
"Cristian",
""
]
] |
2512.10040
|
Skyler Wu
|
Skyler Wu, Aymen Echarghaoui
|
Intelligently Weighting Multiple Reference Models for Direct Preference Optimization of LLMs
|
Working paper. 13 pages, 4 figures
| null | null | null |
cs.LG cs.AI stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Fine-tuning is integral for aligning large language models (LLMs) with human preferences. Multiple-Reference Preference Optimization (MRPO) builds on Direct Preference Optimization (DPO) by fine-tuning LLMs on preference datasets while regularizing the policy towards a mixture of reference models to leverage their collective desirable properties. However, current methods for setting the reference weights are ad-hoc and statistically unsound, leading to unreliable performance. To address this, we introduce four new weighting strategies: two offline methods that leverage held-out validation signal; one online method that uses a sliding-window estimator to reduce overfitting; and an online method that treats reference weighting as a $K$-armed bandit via Thompson Sampling. Experiments using Qwen2.5-0.5B as the policy model and seven reference models from the Llama, Mistral, Qwen, Yi, and Phi families (0.5B-14B each) show that all 4 of our strategies outperform the current MRPO weighting methods on UltraFeedback and SafeRLHF in preference accuracy. More thought-provokingly, however, we find that single-reference DPO, using any of 6 out of 7 references, consistently outperforms all tested multiple-reference approaches -- calling into question the practical appeal of multiple-reference approaches.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:45:20 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wu",
"Skyler",
""
],
[
"Echarghaoui",
"Aymen",
""
]
] |
2512.10041
|
Yihao Liu
|
Yihao Liu, Chenyu Gao, Lianrui Zuo, Michael E. Kim, Brian D. Boyd, Lisa L. Barnes, Walter A. Kukull, Lori L. Beason-Held, Susan M. Resnick, Timothy J. Hohman, Warren D. Taylor, Bennett A. Landman
|
MetaVoxel: Joint Diffusion Modeling of Imaging and Clinical Metadata
| null | null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Modern deep learning methods have achieved impressive results across tasks from disease classification, estimating continuous biomarkers, to generating realistic medical images. Most of these approaches are trained to model conditional distributions defined by a specific predictive direction with a specific set of input variables. We introduce MetaVoxel, a generative joint diffusion modeling framework that models the joint distribution over imaging data and clinical metadata by learning a single diffusion process spanning all variables. By capturing the joint distribution, MetaVoxel unifies tasks that traditionally require separate conditional models and supports flexible zero-shot inference using arbitrary subsets of inputs without task-specific retraining. Using more than 10,000 T1-weighted MRI scans paired with clinical metadata from nine datasets, we show that a single MetaVoxel model can perform image generation, age estimation, and sex prediction, achieving performance comparable to established task-specific baselines. Additional experiments highlight its capabilities for flexible inference.Together, these findings demonstrate that joint multimodal diffusion offers a promising direction for unifying medical AI models and enabling broader clinical applicability.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:47:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Liu",
"Yihao",
""
],
[
"Gao",
"Chenyu",
""
],
[
"Zuo",
"Lianrui",
""
],
[
"Kim",
"Michael E.",
""
],
[
"Boyd",
"Brian D.",
""
],
[
"Barnes",
"Lisa L.",
""
],
[
"Kukull",
"Walter A.",
""
],
[
"Beason-Held",
"Lori L.",
""
],
[
"Resnick",
"Susan M.",
""
],
[
"Hohman",
"Timothy J.",
""
],
[
"Taylor",
"Warren D.",
""
],
[
"Landman",
"Bennett A.",
""
]
] |
2512.10042
|
Meiqi Sun
|
Jongmin Lee, Meiqi Sun, Pieter Abbeel
|
SEMDICE: Off-policy State Entropy Maximization via Stationary Distribution Correction Estimation
|
ICLR 2025
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
In the unsupervised pre-training for reinforcement learning, the agent aims to learn a prior policy for downstream tasks without relying on task-specific reward functions. We focus on state entropy maximization (SEM), where the goal is to learn a policy that maximizes the entropy of the state stationary distribution. In this paper, we introduce SEMDICE, a principled off-policy algorithm that computes an SEM policy from an arbitrary off-policy dataset, which optimizes the policy directly within the space of stationary distributions. SEMDICE computes a single, stationary Markov state-entropy-maximizing policy from an arbitrary off-policy dataset. Experimental results demonstrate that SEMDICE outperforms baseline algorithms in maximizing state entropy while achieving the best adaptation efficiency for downstream tasks among SEM-based unsupervised RL pre-training methods.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:50:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Lee",
"Jongmin",
""
],
[
"Sun",
"Meiqi",
""
],
[
"Abbeel",
"Pieter",
""
]
] |
2512.10043
|
Jo\~ao Lucas Luz Lima Sarcinelli
|
Jo\~ao Lucas Luz Lima Sarcinelli and Diego Furtado Silva
|
Local LLM Ensembles for Zero-shot Portuguese Named Entity Recognition
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) excel in many Natural Language Processing (NLP) tasks through in-context learning but often under-perform in Named Entity Recognition (NER), especially for lower-resource languages like Portuguese. While open-weight LLMs enable local deployment, no single model dominates all tasks, motivating ensemble approaches. However, existing LLM ensembles focus on text generation or classification, leaving NER under-explored. In this context, this work proposes a novel three-step ensemble pipeline for zero-shot NER using similarly capable, locally run LLMs. Our method outperforms individual LLMs in four out of five Portuguese NER datasets by leveraging a heuristic to select optimal model combinations with minimal annotated data. Moreover, we show that ensembles obtained on different source datasets generally outperform individual LLMs in cross-dataset configurations, potentially eliminating the need for annotated data for the current task. Our work advances scalable, low-resource, and zero-shot NER by effectively combining multiple small LLMs without fine-tuning. Code is available at https://github.com/Joao-Luz/local-llm-ner-ensemble.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:55:06 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Sarcinelli",
"João Lucas Luz Lima",
""
],
[
"Silva",
"Diego Furtado",
""
]
] |
2512.10044
|
David Breitenmoser
|
David Breitenmoser, Ricardo Lopez, Shaun D. Clarke, Sara A. Pozzi
|
Identifying Neutron Sources using Recoil and Time-of-Flight Spectroscopy
|
8 pages, 4 figures, 1 ancillary file, submitted to Physical Review Letters
| null | null | null |
physics.ins-det nucl-ex physics.app-ph physics.data-an
|
http://creativecommons.org/licenses/by/4.0/
|
Neutron-source identification is central to nuclear physics and its applications, from planetary science to nuclear security, yet direct discrimination from neutron spectra remains fundamentally elusive. Here, we introduce a Bayesian protocol that directly infers source ensembles from measured neutron spectra by combining full-spectrum template matching with probabilistic evidence evaluation. Applying this protocol to recoil and time-of-flight spectroscopy, we recover single- and two-source configurations with strong statistical significance ($>\!\!4\sigma$) at event counts as low as $\sim\!\!10^{3}$. These results demonstrate that neutron spectral signatures can be leveraged for robust source identification, opening a new observational window for both fundamental research and operationally driven applications.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:57:05 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Breitenmoser",
"David",
""
],
[
"Lopez",
"Ricardo",
""
],
[
"Clarke",
"Shaun D.",
""
],
[
"Pozzi",
"Sara A.",
""
]
] |
2512.10045
|
Helaman Flores
|
Helaman Flores, Mahmoud Jalali Mehrabad, Siavash Mirzaei-Ghormish, Ryan M. Camacho, Dirk Englund
|
Broadband Spatio-Spectral Mode Conversion via Four-Wave Mixing
| null | null | null | null |
quant-ph physics.app-ph physics.optics
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce a framework for scalable and broadband free-space phase-matched four-wave mixing in ring resonators. This method for four-wave mixing reduces the complexity of coupling an emitter to a quantum network by combining the spatial and spectral interfaces between them into one nonlinear optical process. The device is compliant with current heterogeneous integration capabilities and has a bandwidth of 165 nm for efficient spatio-spectral conversion. We outline a fabrication-ready diamond-on-insulator pathway towards modular unit cells that natively bridge visible color centers to the infrared spectrum for scalable quantum networks. We also present and analyze an end-to-end framework for considering single-photon coupling efficiency from a color center to a quantum network. This framework represents a step forwards in analyzing and reducing system-scale losses in a spin-photon interface.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 19:57:05 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Flores",
"Helaman",
""
],
[
"Mehrabad",
"Mahmoud Jalali",
""
],
[
"Mirzaei-Ghormish",
"Siavash",
""
],
[
"Camacho",
"Ryan M.",
""
],
[
"Englund",
"Dirk",
""
]
] |
2512.10046
|
Yan Zhuang
|
Yan Zhuang, Jiawei Ren, Xiaokang Ye, Jianzhi Shen, Ruixuan Zhang, Tianai Yue, Muhammad Faayez, Xuhong He, Ziqiao Ma, Lianhui Qin, Zhiting Hu, Tianmin Shu
|
SimWorld-Robotics: Synthesizing Photorealistic and Dynamic Urban Environments for Multimodal Robot Navigation and Collaboration
|
Conference: NeurIPS 2025 (main)
| null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Recent advances in foundation models have shown promising results in developing generalist robotics that can perform diverse tasks in open-ended scenarios given multimodal inputs. However, current work has been mainly focused on indoor, household scenarios. In this work, we present SimWorld-Robotics~(SWR), a simulation platform for embodied AI in large-scale, photorealistic urban environments. Built on Unreal Engine 5, SWR procedurally generates unlimited photorealistic urban scenes populated with dynamic elements such as pedestrians and traffic systems, surpassing prior urban simulations in realism, complexity, and scalability. It also supports multi-robot control and communication. With these key features, we build two challenging robot benchmarks: (1) a multimodal instruction-following task, where a robot must follow vision-language navigation instructions to reach a destination in the presence of pedestrians and traffic; and (2) a multi-agent search task, where two robots must communicate to cooperatively locate and meet each other. Unlike existing benchmarks, these two new benchmarks comprehensively evaluate a wide range of critical robot capacities in realistic scenarios, including (1) multimodal instructions grounding, (2) 3D spatial reasoning in large environments, (3) safe, long-range navigation with people and traffic, (4) multi-robot collaboration, and (5) grounded communication. Our experimental results demonstrate that state-of-the-art models, including vision-language models (VLMs), struggle with our tasks, lacking robust perception, reasoning, and planning abilities necessary for urban environments.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:04:08 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhuang",
"Yan",
""
],
[
"Ren",
"Jiawei",
""
],
[
"Ye",
"Xiaokang",
""
],
[
"Shen",
"Jianzhi",
""
],
[
"Zhang",
"Ruixuan",
""
],
[
"Yue",
"Tianai",
""
],
[
"Faayez",
"Muhammad",
""
],
[
"He",
"Xuhong",
""
],
[
"Ma",
"Ziqiao",
""
],
[
"Qin",
"Lianhui",
""
],
[
"Hu",
"Zhiting",
""
],
[
"Shu",
"Tianmin",
""
]
] |
2512.10047
|
Zhuo-Yang Song
|
Zhuo-Yang Song, Qing-Hong Cao, Ming-xing Luo, Hua Xing Zhu
|
Detailed balance in large language model-driven agents
|
20 pages, 12 figures, 5 tables
| null | null | null |
cs.LG cond-mat.stat-mech cs.AI nlin.AO physics.data-an
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language model (LLM)-driven agents are emerging as a powerful new paradigm for solving complex problems. Despite the empirical success of these practices, a theoretical framework to understand and unify their macroscopic dynamics remains lacking. This Letter proposes a method based on the least action principle to estimate the underlying generative directionality of LLMs embedded within agents. By experimentally measuring the transition probabilities between LLM-generated states, we statistically discover a detailed balance in LLM-generated transitions, indicating that LLM generation may not be achieved by generally learning rule sets and strategies, but rather by implicitly learning a class of underlying potential functions that may transcend different LLM architectures and prompt templates. To our knowledge, this is the first discovery of a macroscopic physical law in LLM generative dynamics that does not depend on specific model details. This work is an attempt to establish a macroscopic dynamics theory of complex AI systems, aiming to elevate the study of AI agents from a collection of engineering practices to a science built on effective measurements that are predictable and quantifiable.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:04:23 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Song",
"Zhuo-Yang",
""
],
[
"Cao",
"Qing-Hong",
""
],
[
"Luo",
"Ming-xing",
""
],
[
"Zhu",
"Hua Xing",
""
]
] |
2512.10048
|
Michael Greshan Rasiah
|
Michael Greshan Rasiah, Tom J A J Konings, Amanda Nio, Stefano Moriconi, Ashish S Patel, Alberto Smith, Mohamed A Abdelhalim, Tammo Delhaas, M Jorge Cardoso, Pablo Lamata, Barend M E Mees, Bijan Modarai
|
In silico modelling of changes in spinal cord blood flow after endovascular aortic aneurysm repair
| null | null | null | null |
q-bio.TO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Aims: To develop an in-silico model of the aorta and its spinal cord-supplying branches, and to characterise haemodynamic changes following aortic aneurysm (AA) repair with such a model. The work is motivated by the risk of spinal cord ischaemia (SCI) and paraplegia, serious complications that can arise from disruption of spinal cord perfusion during AA surgery. Methods: SimVascular was used to retrospectively create models of a 76 year old female patient's aorta pre- and post- uncomplicated endovascular AA repair. The full extent of the aorta and its branches, including vessels supplying the spinal cord, was segmented. Pulsatile flow simulations were conducted under the assumption of rigid vessel walls, with patient-specific inlet and three-element Windkessel models for the outlet boundary conditions on the SimVascular Gateway Cluster. Results: Postoperatively, segmental artery flow to the spinal cord decreased by 51.86% due to exclusion of lumbar and posterior intercostal arteries by the stent graft. Spinal cord-supplying arteries showed increased TAWSS (+5.2%) and reduced RRT and ECAP, with minimal change in OSI. Consistent with redistribution away from the spinal territory, modest postoperative flow increases were observed in non-spinal vascular beds, including the legs (+6.09%), reno-visceral vessels (+5.89%), and supra-aortic branches (+5.97%). Across vascular territories, visceral arteries had the highest TAWSS and lowest RRT and ECAP, while leg arteries had the lowest TAWSS and highest RRT and ECAP; supra-aortic vessels exhibited the highest OSI. Conclusion: This study lays a foundation for computational prediction of SCI risk. It leverages in-silico modelling, using an open-source software pipeline and routine medical imaging, to assess spinal cord blood flow alterations after aortic surgery. This may forge a path towards a clinical decision-making tool.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:06:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Rasiah",
"Michael Greshan",
""
],
[
"Konings",
"Tom J A J",
""
],
[
"Nio",
"Amanda",
""
],
[
"Moriconi",
"Stefano",
""
],
[
"Patel",
"Ashish S",
""
],
[
"Smith",
"Alberto",
""
],
[
"Abdelhalim",
"Mohamed A",
""
],
[
"Delhaas",
"Tammo",
""
],
[
"Cardoso",
"M Jorge",
""
],
[
"Lamata",
"Pablo",
""
],
[
"Mees",
"Barend M E",
""
],
[
"Modarai",
"Bijan",
""
]
] |
2512.10049
|
Jocelyn Nembe
|
Jocelyn Nembe
|
Adaptive Nonparametric Estimation via Kernel Transport on Group Orbits: Oracle Inequalities and Minimax Rates
| null | null | null | null |
math.ST stat.TH
|
http://creativecommons.org/licenses/by/4.0/
|
We develop a unified framework for nonparametric functional estimation based on kernel transport along orbits of discrete group actions, which we term \emph{Twin Spaces}. Given a base kernel $K$ and a group $G = \langle\varphi\rangle$ acting isometrically on the input space $E$, we construct a hierarchy of transported kernels $\{K_j\}_{j\geq 0}$ and a penalized model selection scheme satisfying a Kraft inequality. Our main contributions are threefold: (i) we establish non-asymptotic oracle inequalities for the penalized twin-kernel estimator with explicit constants; (ii) we introduce novel twin-regularity classes that capture smoothness along group orbits and prove that our estimator adapts to these classes; (iii) we show that the framework recovers classical minimax-optimal rates in the Euclidean setting while enabling improved rates when the target function exhibits orbital structure. The effective dimension $d_{\mathrm{eff}}$ governing the rates is characterized in terms of the quotient $G/L$, where $L$ is the subgroup preserving the base operation. Connections to wavelet methods, geometric quantization, and adaptive computation are discussed.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:07:16 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Nembe",
"Jocelyn",
""
]
] |
2512.10050
|
Christian Millichap
|
Christian Millichap and Rolland Trapp
|
Symmetry groups of flat fully augmented links and their complements
|
24 pages, 13 Figures
| null | null | null |
math.GT
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we prove that the (orientation-preserving) symmetry groups of $b$-prime flat fully augmented links correspond exactly with the finite subgroups of $O(3)$. We accomplish this by first developing a dictionary between automorphisms of a $3$-connected planar cubic graph associated to a flat fully augmented link $L$ and orientation-preserving symmetries of $L$. Our work also provides a simple method to explicitly construct infinite classes of distinct $b$-prime flat fully augmented links $\{L_i\}$ with $Sym^{+}(\mathbb{S}^{3} \setminus L_i) \cong Sym^{+}(\mathbb{S}^{3}, L_i) \cong G$, for any $G$ that is a finite subgroup of $O(3)$.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:11:15 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Millichap",
"Christian",
""
],
[
"Trapp",
"Rolland",
""
]
] |
2512.10051
|
Moulik Gupta
|
Moulik Gupta, Achyut Mani Tripathi
|
DB2-TransF: All You Need Is Learnable Daubechies Wavelets for Time Series Forecasting
| null | null | null | null |
cs.LG cs.AI eess.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Time series forecasting requires models that can efficiently capture complex temporal dependencies, especially in large-scale and high-dimensional settings. While Transformer-based architectures excel at modeling long-range dependencies, their quadratic computational complexity poses limitations on scalability and adaptability. To overcome these challenges, we introduce DB2-TransF, a novel Transformer-inspired architecture that replaces the self-attention mechanism with a learnable Daubechies wavelet coefficient layer. This wavelet-based module efficiently captures multi-scale local and global patterns and enhances the modeling of correlations across multiple time series for the time series forecasting task. Extensive experiments on 13 standard forecasting benchmarks demonstrate that DB2-TransF achieves comparable or superior predictive accuracy to conventional Transformers, while substantially reducing memory usage for the time series forecasting task. The obtained experimental results position DB2-TransF as a scalable and resource-efficient framework for advanced time series forecasting. Our code is available at https://github.com/SteadySurfdom/DB2-TransF
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:15:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gupta",
"Moulik",
""
],
[
"Tripathi",
"Achyut Mani",
""
]
] |
2512.10052
|
Nestor Emilio Massa
|
N\'estor E. Massa, Javier Gainza, Aur\'elien Canizares, Leire del Campo, and Jos\'e Antonio Alonso
|
Spin-phonon interactions revisited: Far-infrared emission, Raman scattering, and high-resolution x-ray diffraction at the N\'eel temperature in LaFeO3
|
10 figures
|
Phys. Rev. B 112, 214420 (2025)
|
10.1103/vygw-lm64
| null |
cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by/4.0/
|
We report on the structural evolution, spin phonon interactions, and magnetoelastic effects in bulk LaFeO3 perovskite across its antiferromagnetic transition. We found that while the lattice constants exhibit slight deviations at the Neel temperature(TN), due to spin phonon interactions, individual interatomic bonds display distinct behavior. The octahedral B site basal plane, involving oxygen ion pairs that mediate superexchange interactions, displays bond contraction (or elongation) assimilable to the observed in x ray diffraction for negative thermal expansion. These bond deviations reach a singular point near TN, where a sharp change in bond length is observed. A similar abrupt change is also detected in the nearest neighbor La O2 basal plane distances, which we interpret as arising from ion size differences within the 3d 2p hybridized states associated with the superexchange mechanism. High-resolution SXRD measurements of the c-axis Fe3+ O1 and La3+ O1 apex distances reveal a distinct inflection at the Neel temperature. This behavior supports the role of lattice distortions and crystal field modifications in the emergence of non collinear weak ferromagnetism in RFeO3 (R = rare earth) perovskites. Furthermore, our SXRD patterns show an increase in diffraction line intensities peaking at TN, suggesting a straightforward indicator for spin-phonon interactions that also resolves into lattice metastability deduced by nonlinear thermal expansion in the paramagnetic phase. Preliminary results across the full RFeO3 (R = rare earth) series indicate a shared structural framework among all members, potentially offering insights into previously inconclusive structural interpretations in oxides with common octahedral sublattices.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:18:05 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Massa",
"Néstor E.",
""
],
[
"Gainza",
"Javier",
""
],
[
"Canizares",
"Aurélien",
""
],
[
"del Campo",
"Leire",
""
],
[
"Alonso",
"José Antonio",
""
]
] |
2512.10053
|
Mario Figueiredo
|
Tiago Brogueira, M\'ario A. T. Figueiredo
|
LxCIM: a new rank-based binary classifier performance metric invariant to local exchange of classes
|
28 pages, 7 figures
| null | null | null |
stat.ML cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Binary classification is one of the oldest, most prevalent, and studied problems in machine learning. However, the metrics used to evaluate model performance have received comparatively little attention. The area under the receiver operating characteristic curve (AUROC) has long been a standard choice for model comparison. Despite its advantages, AUROC is not always ideal, particularly for problems that are invariant to local exchange of classes (LxC), a new form of metric invariance introduced in this work. To address this limitation, we propose LxCIM (LxC-invariant metric), which is not only rank-based and invariant under local exchange of classes, but also intuitive, logically consistent, and always computable, while enabling more detailed analysis through the cumulative accuracy-decision rate curve. Moreover, LxCIM exhibits clear theoretical connections to AUROC, accuracy, and the area under the accuracy-decision rate curve (AUDRC). These relationships allow for multiple complementary interpretations: as a symmetric form of AUROC, a rank-based analogue of accuracy, or a more representative and more interpretable variant of AUDRC. Finally, we demonstrate the direct applicability of LxCIM to the bivariate causal discovery problem (which exhibits invariance to local exchange of classes) and show how it addresses the acknowledged limitations of existing metrics used in this field. All code and implementation details are publicly available at github.com/tiagobrogueira/Causal-Discovery-In-Exchangeable-Data.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:18:19 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Brogueira",
"Tiago",
""
],
[
"Figueiredo",
"Mário A. T.",
""
]
] |
2512.10054
|
Logan Robbins
|
Logan Robbins
|
Parallel Decoder Transformer: Model-Internal Parallel Decoding with Speculative Invariance via Note Conditioning
| null | null | null | null |
cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Autoregressive decoding in Large Language Models (LLMs) is inherently sequential, creating a latency bottleneck that scales linearly with output length. While ``Decomposition-and-Fill'' methods like Skeleton-of-Thought attempt to parallelize generation via external orchestration, they suffer from \textit{coherence drift} due to the lack of cross-stream communication. In this work, we introduce the \textbf{Parallel Decoder Transformer (PDT)}, a parameter-efficient architecture that embeds coordination primitives directly into the inference process of a frozen pre-trained model.
Instead of retraining the base model, PDT injects lightweight \textit{Speculative Note Conditioning (SNC)} adapters that allow parallel decoding streams to synchronize via a shared, dynamic latent space. We formulate coordination as a \textit{speculative consensus} problem, where sibling streams broadcast semantic ``notes'' to a global bus, gated by a learned verification head. We validate our approach on a 50,000-step curriculum using a frozen 20B-parameter backbone. Our results demonstrate that PDT achieves effective self-correction, reaching \textbf{77.8\% precision} in coverage prediction and recovering approximate serial semantics without modifying the trunk weights. This establishes PDT as a scalable, efficient alternative to full model fine-tuning for structured parallel generation.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:19:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Robbins",
"Logan",
""
]
] |
2512.10055
|
Yannick Kuhn
|
Yannick Kuhn, Masaki Adachi, Micha Philipp, David A. Howey, Birger Horstmann
|
A Primer on Bayesian Parameter Estimation and Model Selection for Battery Simulators
|
22 pages, 19 figures
| null | null | null |
stat.ME physics.data-an stat.AP
|
http://creativecommons.org/licenses/by/4.0/
|
Physics-based battery modelling has emerged to accelerate battery materials discovery and performance assessment. Its success, however, is still hindered by difficulties in aligning models to experimental data. Bayesian approaches are a valuable tool to overcome these challenges, since they enable prior assumptions and observations to be combined in a principled manner that improves numerical conditioning. Here we introduce two new algorithms to the battery community, SOBER and BASQ, that greatly speed up Bayesian inference for parameterisation and model comparison. We showcase how Bayesian model selection allows us to tackle data observability, model identifiability, and data-informed model development together. We propose this approach for the search for battery models of novel materials.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:21:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kuhn",
"Yannick",
""
],
[
"Adachi",
"Masaki",
""
],
[
"Philipp",
"Micha",
""
],
[
"Howey",
"David A.",
""
],
[
"Horstmann",
"Birger",
""
]
] |
2512.10056
|
Alireza Namazi
|
Alireza Namazi, Amirreza Dolatpour Fathkouhi, Heman Shakeri
|
Mitigating Exposure Bias in Risk-Aware Time Series Forecasting with Soft Tokens
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Autoregressive forecasting is central to predictive control in diabetes and hemodynamic management, where different operating zones carry different clinical risks. Standard models trained with teacher forcing suffer from exposure bias, yielding unstable multi-step forecasts for closed-loop use. We introduce Soft-Token Trajectory Forecasting (SoTra), which propagates continuous probability distributions (``soft tokens'') to mitigate exposure bias and learn calibrated, uncertainty-aware trajectories. A risk-aware decoding module then minimizes expected clinical harm. In glucose forecasting, SoTra reduces average zone-based risk by 18\%; in blood-pressure forecasting, it lowers effective clinical risk by approximately 15\%. These improvements support its use in safety-critical predictive control.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:25:45 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Namazi",
"Alireza",
""
],
[
"Fathkouhi",
"Amirreza Dolatpour",
""
],
[
"Shakeri",
"Heman",
""
]
] |
2512.10057
|
Jiahao Jiang
|
Jiahao Jiang
|
Towards a Mathematical Theory of Adaptive Memory: From Time-Varying to Responsive Fractional Brownian Motion
|
108 pages, 0 figures. Submitted to arXiv
| null | null | null |
math.PR math-ph math.MP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This work develops a comprehensive mathematical theory for a class of stochastic processes whose local regularity adapts dynamically in response to their own state. We first introduce and rigorously analyze a time-varying fractional Brownian motion (TV-fBm) with a deterministic, H\"older-continuous Hurst exponent function. Key properties are established, including its exact variance scaling law, precise local increment asymptotics, local non-determinism, large deviation asymptotics for its increments, and a covariance structure that admits a closed-form hypergeometric representation. We then define a novel class of processes termed Responsive Fractional Brownian Motion (RfBm). Here,the Hurst exponent is governed by a Lipschitz-H\"older response function depending on the process state itself, creating an intrinsic feedback mechanism between state and memory. We establish the well-posedness of this definition, prove pathwise H\"older regularity of the induced instantaneous scaling exponent, and analyze associated cumulative memory processes along with their asymptotic convergence. The mathematical structure of RfBm naturally gives rise to a continuous-time, pathwise attention mechanism. We show that its kernel induces a well-defined attention weight distribution, derive fundamental bounds for these weights, and quantify the stability of attentional allocation through residence measures and volatility functionals. This work develops a stochastic-process-theoretic framework for concepts central to adaptive memory and content-sensitive information processing, offering a mathematically grounded perspective that may complement existing empirical approaches.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:26:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Jiang",
"Jiahao",
""
]
] |
2512.10058
|
Daniel Roytburg
|
Dani Roytburg, Beck Miller
|
Mind the Gap! Pathways Towards Unifying AI Safety and Ethics Research
|
Accepted for presentation at IASEAI 2026
| null | null | null |
cs.AI cs.CY cs.HC cs.SI
|
http://creativecommons.org/licenses/by/4.0/
|
While much research in artificial intelligence (AI) has focused on scaling capabilities, the accelerating pace of development makes countervailing work on producing harmless, "aligned" systems increasingly urgent. Yet research on alignment has diverged along two largely parallel tracks: safety--centered on scaled intelligence, deceptive or scheming behaviors, and existential risk--and ethics--focused on present harms, the reproduction of social bias, and flaws in production pipelines. Although both communities warn of insufficient investment in alignment, they disagree on what alignment means or ought to mean. As a result, their efforts have evolved in relative isolation, shaped by distinct methodologies, institutional homes, and disciplinary genealogies.
We present a large-scale, quantitative study showing the structural split between AI safety and AI ethics. Using a bibliometric and co-authorship network analysis of 6,442 papers from twelve major ML and NLP conferences (2020-2025), we find that over 80% of collaborations occur within either the safety or ethics communities, and cross-field connectivity is highly concentrated: roughly 5% of papers account for more than 85% of bridging links. Removing a small number of these brokers sharply increases segregation, indicating that cross-disciplinary exchange depends on a handful of actors rather than broad, distributed collaboration. These results show that the safety-ethics divide is not only conceptual but institutional, with implications for research agendas, policy, and venues. We argue that integrating technical safety work with normative ethics--via shared benchmarks, cross-institutional venues, and mixed-method methodologies--is essential for building AI systems that are both robust and just.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:28:13 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Roytburg",
"Dani",
""
],
[
"Miller",
"Beck",
""
]
] |
2512.10059
|
Rasmus Vikhamar-Sandberg
|
Rasmus Vikhamar-Sandberg, Michal Repisky
|
Efficient Boys function evaluation using minimax approximation
| null | null | null | null |
math.NA cs.NA physics.comp-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present an algorithm for efficient evaluation of Boys functions $F_0,\dots,F_{k_\mathrm{max}}$ tailored to modern computing architectures, in particular graphical processing units (GPUs), where maximum throughput is high and data movement is costly. The method combines rational minimax approximations with upward and downward recurrence relations. The non-negative real axis is partitioned into three regions, $[0,\infty\rangle = A\cup B\cup C$, where regions $A$ and $B$ are treated using rational minimax approximations and region $C$ by an asymptotic approximation. This formulation avoids lookup tables and irregular memory access, making it well suited hardware with high maximum throughput and low latency. The rational minimax coefficients are generated using the rational Remez algorithm. For a target maximum absolute error of $\varepsilon_\mathrm{tol} = 5\cdot10^{-14}$, the corresponding approximation regions and coefficients for Boys functions $F_0,\dots,F_{32}$ are provided in the appendix.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:28:14 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Vikhamar-Sandberg",
"Rasmus",
""
],
[
"Repisky",
"Michal",
""
]
] |
2512.10060
|
Brandt Gaches
|
Brandt A. L. Gaches and Serena Viti
|
High-energy astrochemistry in the molecular interstellar medium
|
Accepted by ACS Earth and Space Chemistry (Eric Herbst Festschrift)
| null | null | null |
astro-ph.GA astro-ph.EP astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In the past decade, there has been a significant shift in astrochemistry with a renewed focus on the role of non-thermal processes on the molecular interstellar medium, in particular energetic particles (such as cosmic ray particles and fast electrons) and X-ray radiation. This has been brought about in large part due to new observations of interstellar complex organic molecules (iCOMS) in environments that would inhibit their formation, such as cold, dense gas in prestellar cores or in the highly energetic environments in galactic centers. In parallel, there has been a plethora of new laboratory investigations on the role of high-energy radiation and electrons on the chemistry of astrophysical ices, demonstrating the ability of this radiation to induce complex chemistry. In recent years, theoretical models have also begun to include newer cosmic-ray-driven processes in both the gas and ice phases. In this review, we unify aspects of the chemistry driven by X-ray radiation and energetic particles into a ``high-energy astrochemistry'', defining this term and reviewing the underlying chemical processes. We conclude by examining various laboratories where high-energy astrochemistry is at play and identify future issues to be tackled.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:28:28 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gaches",
"Brandt A. L.",
""
],
[
"Viti",
"Serena",
""
]
] |
2512.10061
|
Jo\~ao Lucas Luz Lima Sarcinelli
|
Jo\~ao Lucas Luz Lima Sarcinelli and Ricardo Marcondes Marcacini
|
\textsc{Text2Graph}: Combining Lightweight LLMs and GNNs for Efficient Text Classification in Label-Scarce Scenarios
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) have become effective zero-shot classifiers, but their high computational requirements and environmental costs limit their practicality for large-scale annotation in high-performance computing (HPC) environments. To support more sustainable workflows, we present \textsc{Text2Graph}, an open-source Python package that provides a modular implementation of existing text-to-graph classification approaches. The framework enables users to combine LLM-based partial annotation with Graph Neural Network (GNN) label propagation in a flexible manner, making it straightforward to swap components such as feature extractors, edge construction methods, and sampling strategies. We benchmark \textsc{Text2Graph} on a zero-shot setting using five datasets spanning topic classification and sentiment analysis tasks, comparing multiple variants against other zero-shot approaches for text classification. In addition to reporting performance, we provide detailed estimates of energy consumption and carbon emissions, showing that graph-based propagation achieves competitive results at a fraction of the energy and environmental cost.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:31:30 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Sarcinelli",
"João Lucas Luz Lima",
""
],
[
"Marcacini",
"Ricardo Marcondes",
""
]
] |
2512.10062
|
Max Weinreich
|
Max Weinreich
|
Degree growth of skew pentagram maps
|
34 pages
| null | null | null |
math.DS
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Skew pentagram maps act on polygons by intersecting diagonals of different lengths. They were introduced by Khesin-Soloviev in 2015 as conjecturally non-integrable generalizations of the pentagram map, a well-known integrable system. In this paper, we show that certain skew pentagram maps have exponential degree growth and no preserved fibration. To formalize this, we introduce a general notion of first dynamical degree for lattice maps, or shift-invariant self-maps of $(\mathbb{P}^N)^\mathbb{Z}$. We show that the dynamical degree of any equal-length pentagram map is 1, but that there are infinitely many skew pentagram maps with dynamical degree 4.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:32:30 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Weinreich",
"Max",
""
]
] |
2512.10063
|
Ravi Kunjwal
|
Ravi Kunjwal
|
Quantumness via Discrete Structures
|
HDR dissertation defended on December 8, 2025, at Aix-Marseille University
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Quantum theory departs from classical probabilistic theories in foundational ways. These departures--termed quantumness here--power quantum information and computation. This thesis charts the role of discrete structures in assessing quantumness, synthesizing elements of my postdoctoral research through this lens. After an introduction to the necessary background concepts, I present my work under three broad categories. First, I present work on contextuality that extensively relies on (undirected) graphs and hypergraphs as the discrete structures of interest; more specifically, it relies on invariants associated with them. This work includes Kochen-Specker (KS) contextuality and its operationalization to generalized contextuality, expressed via (hyper)graph-theoretic frameworks. I also present work on KS-contextuality in multiqubit systems and an application of generalized contextuality to a one-shot communication task, both of which rely on hypergraphs. Second, I present work on causality, where the discrete structures of interest are directed graphs. This includes work on indefinite causal order, specifically its connections to the gap between local operations and classical communication (LOCC) and separable operations (SEP), and a device-independent notion of nonclassicality--termed antinomicity--that generalizes Bell nonlocality without global causal assumptions. Finally, I present work on the incompatibility of quantum measurements, its connection to Bell nonlocality, and its role in discriminating between quantum and almost quantum correlations in the single-system setting. The discrete structures of interest here are hypergraphs that model joint measurability relations between quantum measurements. I conclude with a summary and an overview of work that is not covered in this thesis.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:34:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kunjwal",
"Ravi",
""
]
] |
2512.10064
|
Samuel Mimram
|
Samuel Mimram and \'Emile Oleon
|
Classifying covering types in homotopy type theory
| null | null | null | null |
cs.LO math.AT
|
http://creativecommons.org/licenses/by/4.0/
|
Covering spaces are a fundamental tool in algebraic topology because of the close relationship they bear with the fundamental groups of spaces. Indeed, they are in correspondence with the subgroups of the fundamental group: this is known as the Galois correspondence. In particular, the covering space corresponding to the trivial group is the universal covering, which is a "1-connected" variant of the original space, in the sense that it has the same homotopy groups, except for the first one which is trivial. In this article, we formalize this correspondence in homotopy type theory, a variant of Martin-L\"of type theory in which types can be interpreted as spaces (up to homotopy). Along the way, we develop an n-dimensional generalization of covering spaces. Moreover, in order to demonstrate the applicability of our approach, we formally classify the covering of lens spaces and explain how to construct the Poincar\'e homology sphere.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:34:41 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mimram",
"Samuel",
""
],
[
"Oleon",
"Émile",
""
]
] |
2512.10065
|
Pedro Ramaciotti
|
Paul Bouchaud, Pedro Ramaciotti
|
Linear socio-demographic representations emerge in Large Language Models from indirect cues
| null | null | null | null |
cs.AI cs.HC
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We investigate how LLMs encode sociodemographic attributes of human conversational partners inferred from indirect cues such as names and occupations. We show that LLMs develop linear representations of user demographics within activation space, wherein stereotypically associated attributes are encoded along interpretable geometric directions. We first probe residual streams across layers of four open transformer-based LLMs (Magistral 24B, Qwen3 14B, GPT-OSS 20B, OLMo2-1B) prompted with explicit demographic disclosure. We show that the same probes predict demographics from implicit cues: names activate census-aligned gender and race representations, while occupations trigger representations correlated with real-world workforce statistics. These linear representations allow us to explain demographic inferences implicitly formed by LLMs during conversation. We demonstrate that these implicit demographic representations actively shape downstream behavior, such as career recommendations. Our study further highlights that models that pass bias benchmark tests may still harbor and leverage implicit biases, with implications for fairness when applied at scale.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:36:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bouchaud",
"Paul",
""
],
[
"Ramaciotti",
"Pedro",
""
]
] |
2512.10066
|
Yongkai Chen
|
Yongkai Chen, Samuel WK Wong, SC Kou
|
Classifying Metamorphic versus Single-Fold Proteins with Statistical Learning and AlphaFold2
| null | null | null | null |
stat.AP cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
The remarkable success of AlphaFold2 in providing accurate atomic-level prediction of protein structures from their amino acid sequence has transformed approaches to the protein folding problem. However, its core paradigm of mapping one sequence to one structure may only be appropriate for single-fold proteins with one stable conformation. Metamorphic proteins, which can adopt multiple distinct conformations, have conformational diversity that cannot be adequately modeled by AlphaFold2. Hence, classifying whether a given protein is metamorphic or single-fold remains a critical challenge for both laboratory experiments and computational methods. To address this challenge, we developed a novel classification framework by re-purposing AlphaFold2 to generate conformational ensembles via a multiple sequence alignment sampling method. From these ensembles, we extract a comprehensive set of features characterizing the conformational ensemble's modality and structural dispersion. A random forest classifier trained on a carefully curated benchmark dataset of known metamorphic and single-fold proteins achieves a mean AUC of 0.869 with cross-validation, demonstrating the effectiveness of our integrated approach. Furthermore, by applying our classifier to 600 randomly sampled proteins from the Protein Data Bank, we identified several potential metamorphic protein candidates -- including the 40S ribosomal protein S30, whose conformational change is crucial for its secondary function in antimicrobial defense. By combining AI-driven protein structure prediction with statistical learning, our work provides a powerful new approach for discovering metamorphic proteins and deepens our understanding of their role in their molecular function.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:37:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Yongkai",
""
],
[
"Wong",
"Samuel WK",
""
],
[
"Kou",
"SC",
""
]
] |
2512.10068
|
Jocelyn Nembe
|
Jocelyn Nemb\'e
|
TwinKernel Estimation for Point Process Intensity Functions: Adaptive Nonparametric Methods via Orbital Regularity
| null | null | null | null |
math.ST stat.TH
|
http://creativecommons.org/licenses/by/4.0/
|
We develop TwinKernel methods for nonparametric estimation of intensity functions of point processes. Building on the general TwinKernel framework and combining it with martingale techniques for counting processes, we construct estimators that adapt to orbital regularity of the intensity function. Given a point process $N$ with intensity $\lambda$ and a cyclic group $G = \langle\varphi\rangle$ acting on the time/space domain, we transport kernels along group orbits to create a hierarchy of smoothed Nelson-Aalen type estimators. Our main results establish: (i) uniform consistency via martingale concentration inequalities; (ii) optimal convergence rates for intensities in twin-H\"older classes, with rates depending on the effective dimension $d_{\mathrm{eff}}$; (iii) adaptation to unknown smoothness through penalized model selection; (iv) automatic boundary bias correction via local polynomial extensions in twin coordinates; (v) minimax lower bounds showing rate optimality. We apply the methodology to hazard rate estimation under random censoring, where periodicity or other orbital structure in the hazard may arise from circadian rhythms, seasonal effects, or treatment schedules. Martingale central limit theorems yield asymptotic confidence bands. Simulation studies demonstrate 3--7$\times$ improvements over classical kernel hazard estimators when the intensity exhibits orbital regularity.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:43:30 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Nembé",
"Jocelyn",
""
]
] |
2512.10069
|
Chloe Si
|
Chloe Si, David A. Stephens, Erica E.M. Moodie
|
Incorporating Partial Adherence for Estimation of Dynamic Treatment Regimes
| null | null | null | null |
stat.ME
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Dynamic Treatment Regimes (DTRs) provide a systematic framework for optimizing sequential decision-making in chronic disease management, where therapies must adapt to patients' evolving clinical profiles. Inverse probability weighting (IPW) is a cornerstone methodology for estimating regime values from observational data due to its intuitive formulation and established theoretical properties, yet standard IPW estimators face significant limitations, including variance instability and data inefficiency. A fundamental but underexplored source of inefficiency lies in the strict binary adherence criterion that fails to account for partial adherence, thereby discarding substantial data from individuals with even minimal deviations from the target regime. We propose two novel methodologies that relax the strict inclusion rule through flexible compatibility mechanisms. Both methods provide computationally tractable alternatives that can be easily integrated into existing IPW workflows, offering more efficient approaches to DTR estimation. Theoretical analysis demonstrates that both estimators preserve consistency while achieving superior finite-sample efficiency compared to standard IPW, and comprehensive simulation studies confirm improved stability. We illustrate the practical utility of our methods through an application to HIV treatment data from the AIDS Clinical Trials Group Study 175 (ACTG175).
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:43:40 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Si",
"Chloe",
""
],
[
"Stephens",
"David A.",
""
],
[
"Moodie",
"Erica E. M.",
""
]
] |
2512.10070
|
Mark Stephen Senn
|
Ben R. M. Tragheim, Fabio Orlandi, En-Pei Liu, Wei-Tin Chen, Mark S. Senn
|
The magnetic structure of polar $G$-type charge and orbital ordered Hg-quadruple manganite perovskites
|
9 page, 6 figures
| null | null | null |
cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
The magnetic structure of the novel Hg$_{0.7}$Na$_{0.3}$Mn$_3$Mn$_4$O$_{12}$, a quadruple manganite perovskite that exhibits a unique $G$-type charge and orbital ordered state distinct to other $A^{2+}$Mn$_3$Mn$_4$O$_{12}$ equivalents ($A$ $=$ Ca, Sr, Cd, Pb), has been solved using powder neutron diffraction and symmetry-motivated analysis. A $G$-type-like antiferromagnetic (AFM) ordering of Mn on the $A'$ sites and a `up--up--down--down' AFM moment configuration of Mn spins on the $B$ sites is found to occur. The mechanism for the onset and stabilization of $B$ site `up--up--down--down' AFM order is explored in terms of coupling between structural and magnetic distortions. The results presented here provide evidence of the exotic charge, orbital, electronic and magnetic orderings that quadruple manganite perovskites demonstrate, and further highlighting the distinct chemistry that Hg$^{2+}$ plays in stabilizing novel states compared to other divalent $A$-site cation equivalents.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:44:17 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tragheim",
"Ben R. M.",
""
],
[
"Orlandi",
"Fabio",
""
],
[
"Liu",
"En-Pei",
""
],
[
"Chen",
"Wei-Tin",
""
],
[
"Senn",
"Mark S.",
""
]
] |
2512.10071
|
Zhaoshuo Li
|
Junjie Bai, Yu-Wei Chao, Qizhi Chen, Jinwei Gu, Moo Jin Kim, Zhaoshuo Li, Xuan Li, Tsung-Yi Lin, Ming-Yu Liu, Nic Ma, Kaichun Mo, Delin Qu, Shangkun Sun, Hongchi Xia, Fangyin Wei, Xiaohui Zeng
|
Openpi Comet: Competition Solution For 2025 BEHAVIOR Challenge
|
preprint
| null | null | null |
cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
The 2025 BEHAVIOR Challenge is designed to rigorously track progress toward solving long-horizon tasks by physical agents in simulated environments. BEHAVIOR-1K focuses on everyday household tasks that people most want robots to assist with and these tasks introduce long-horizon mobile manipulation challenges in realistic settings, bridging the gap between current research and real-world, human-centric applications. This report presents our solution to the 2025 BEHAVIOR Challenge in a very close 2nd place and substantially outperforms the rest of the submissions. Building on $\pi_{0.5}$, we focus on systematically building our solution by studying the effects of training techniques and data. Through careful ablations, we show the scaling power in pre-training and post-training phases for competitive performance. We summarize our practical lessons and design recommendations that we hope will provide actionable insights for the broader embodied AI community when adapting powerful foundation models to complex embodied scenarios.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:46:40 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bai",
"Junjie",
""
],
[
"Chao",
"Yu-Wei",
""
],
[
"Chen",
"Qizhi",
""
],
[
"Gu",
"Jinwei",
""
],
[
"Kim",
"Moo Jin",
""
],
[
"Li",
"Zhaoshuo",
""
],
[
"Li",
"Xuan",
""
],
[
"Lin",
"Tsung-Yi",
""
],
[
"Liu",
"Ming-Yu",
""
],
[
"Ma",
"Nic",
""
],
[
"Mo",
"Kaichun",
""
],
[
"Qu",
"Delin",
""
],
[
"Sun",
"Shangkun",
""
],
[
"Xia",
"Hongchi",
""
],
[
"Wei",
"Fangyin",
""
],
[
"Zeng",
"Xiaohui",
""
]
] |
2512.10072
|
Megan Delamer
|
Megan Delamer (1 and 2), Gu{\dh}mundur Stef\'ansson (3), Suvrath Mahadevan (1 and 2), Caleb I. Ca\~nas (4), Harish K. Vedantham (5 and 6), Arvind F. Gupta (7), Joseph R. Callingham (5 and 3), Juan Bautista Climent Oliver (8), William Cochran (9 and 10), Rachel B. Fernandes (1 and 2), Evan Fitzmaurice (1 and 2), Jose Carlos Guirado (8 and 11), Michael Hartmann (12), Artie P. Hatzes (12), Elise Koo (3), Jessica E. Libby-Roberts (1 and 2), Joe P. Ninan (13), Miguel P\'erez-Torres (14 and 15), Paul Robertson (16), Arpita Roy (17), Christian Schwab (18) ((1) Department of Astronomy and Astrophysics Pennsylvania State University, (2) Center for Exoplanets and Habitable Worlds, (3) Anton Pannekoek Institute for Astronomy University of Amsterdam, (4) NASA Goddard Space Flight Center, (5) ASTRON Netherlands Institute for Radio Astronomy, (6) Kapteyn Astronomical Institute University of Groninngen, (7) US National Science Foundation National Optical-Infrared Astronomy Research Laboratory, (8) Departament d'Astronomia i Astrofisica Universitat de Valencia, (9) McDonald Observatory, (10) Center for Planetary Systems Habitability, (11) Observatori Astron\`omic Universitat de Val\`encia, (12) Th\"uringer Landessternwarte Tautenburg, (13) Department of Astronomy and Astrophysics Tata Institute of Fundamental Research, (14) Instituto de Astrof\'isica de Andaluc\'ia, (15) School of Sciences European University Cyprus, (16) Department of Physics & Astronomy The University of California Irvine, (17) Astrophysics & Space Institute Schmidt Sciences, (18) School of Mathematical and Physical Sciences Macquarie University)
|
Hipparcos, Gaia, and RVs reveal that the radio emitting F star HD 220242 has an M dwarf companion, a likely source of the radio emission
|
12 pages, 5 figures, Accepted to The Astronomical Journal
| null | null | null |
astro-ph.SR astro-ph.EP
|
http://creativecommons.org/licenses/by/4.0/
|
The detection of circularly polarized, low frequency radio emission offers the tantalizing possibility of the observation of interactions between stars and their possible substellar companions, as well as direct emission from exoplanets. Additional follow up of systems with radio emission is key to understanding the true origin of the emission, since multiple astrophysical mechanisms can plausibly lead to such signals. While nineteen M dwarfs were detected by LOFAR in circular polarization as part of the V-LoTSS survey, HD~220242 is the only F star to have a circularly polarized low frequency radio detection in the same survey. We conducted radial velocity follow up with the Habitable-zone Planet Finder and combined these observations with additional archival RVs and \textit{Hipparcos}-\textit{Gaia} proper motion accelerations to determine that HD~220242 has a stellar companion with P=16.79$\pm$0.04\,yrs and a mass of $0.619\pm0.014$\,M$_\odot$. We use Spectral Energy Distribution fitting and lack of any UV excess to rule out a co-evolved white dwarf companion and confirm that the companion is an M dwarf star. Given that F stars lack the coronal properties to produce such coherent emission, and the companion mass and lack of UV excess are consistent with an M dwarf, the radio emission is most plausibly associated with the companion.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:48:14 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Delamer",
"Megan",
"",
"1 and 2"
],
[
"Stefánsson",
"Guðmundur",
"",
"Anton Pannekoek Institute for Astronomy University of Amsterdam"
],
[
"Mahadevan",
"Suvrath",
"",
"1 and 2"
],
[
"Cañas",
"Caleb I.",
"",
"NASA Goddard Space Flight Center"
],
[
"Vedantham",
"Harish K.",
"",
"5 and 6"
],
[
"Gupta",
"Arvind F.",
"",
"US National Science Foundation National Optical-Infrared Astronomy Research Laboratory"
],
[
"Callingham",
"Joseph R.",
"",
"5 and 3"
],
[
"Oliver",
"Juan Bautista Climent",
"",
"Departament d'Astronomia i Astrofisica Universitat de Valencia"
],
[
"Cochran",
"William",
"",
"9 and 10"
],
[
"Fernandes",
"Rachel B.",
"",
"1 and 2"
],
[
"Fitzmaurice",
"Evan",
"",
"1 and 2"
],
[
"Guirado",
"Jose Carlos",
"",
"8 and 11"
],
[
"Hartmann",
"Michael",
"",
"Thüringer Landessternwarte Tautenburg"
],
[
"Hatzes",
"Artie P.",
"",
"Thüringer Landessternwarte Tautenburg"
],
[
"Koo",
"Elise",
"",
"Anton Pannekoek Institute for Astronomy University of Amsterdam"
],
[
"Libby-Roberts",
"Jessica E.",
"",
"1 and 2"
],
[
"Ninan",
"Joe P.",
"",
"Department of Astronomy and Astrophysics Tata Institute of Fundamental Research"
],
[
"Pérez-Torres",
"Miguel",
"",
"14 and 15"
],
[
"Robertson",
"Paul",
"",
"Department of Physics & Astronomy The University of California Irvine"
],
[
"Roy",
"Arpita",
"",
"Astrophysics & Space Institute Schmidt Sciences"
],
[
"Schwab",
"Christian",
"",
"School of Mathematical and Physical Sciences Macquarie University"
]
] |
2512.10073
|
Stephan Reitzenstein
|
Imad Limame, Peter Ludewig, Aris Koulas-Simos, Chirag C. Palekar, Jan Donges, Ching-Wen Shih, Kartik Gaur, Sarthak Tripathi, Sven Rodt, Wolfgang Stolz, Kerstin Volz, Stephan Reitzenstein
|
Direct Epitaxial Growth and Deterministic Device Integration of high-quality Telecom O-Band InGaAs Quantum Dots on Silicon Substrate
| null | null | null | null |
cond-mat.mes-hall
|
http://creativecommons.org/licenses/by/4.0/
|
Semiconductor quantum dots (QDs) are key building blocks for photonic quantum technologies, enabling practical sources of non-classical light. A central challenge for scalable integration is the direct epitaxial growth of high-quality emitters on industry-compatible silicon platforms. Furthermore, for long-distance fiber-based quantum communication, emission in the telecom O- or C-band is essential. Here, we demonstrate the direct growth of high-quality InGaAs/GaAs QDs emitting in the telecom O-band using a strain-reducing layer approach on silicon. Deterministic integration of individual QDs into circular Bragg grating resonators is achieved via in-situ electron-beam lithography. The resulting devices exhibit strong out-coupling enhancement, with photon extraction efficiencies up to $(40 \pm 2)\%$, in excellent agreement with numerical simulations. These results highlight the high material quality of both the epitaxial platform and the photonic nanostructure, as well as the precise lateral positioning of the emitter within 20~nm of the resonator center. At cryogenic temperature (4~K) and low excitation power ($0.027\times P_\text{sat}$), the devices show excellent single-photon purity, exceeding 99\%. Operation at elevated temperatures of 40~K and 77~K, compatible with compact Stirling cryo-coolers and liquid-nitrogen cooling, reveals robust performance, with single-photon purity maintained at $(88.4 \pm 0.6)\%$ at 77~K. These results demonstrate a practical and scalable route toward silicon-based quantum light sources and provide a promising path for cost-effective fabrication and seamless integration of quantum photonics with classical electronics, representing an important step toward large-scale, chip-based quantum information systems.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:48:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Limame",
"Imad",
""
],
[
"Ludewig",
"Peter",
""
],
[
"Koulas-Simos",
"Aris",
""
],
[
"Palekar",
"Chirag C.",
""
],
[
"Donges",
"Jan",
""
],
[
"Shih",
"Ching-Wen",
""
],
[
"Gaur",
"Kartik",
""
],
[
"Tripathi",
"Sarthak",
""
],
[
"Rodt",
"Sven",
""
],
[
"Stolz",
"Wolfgang",
""
],
[
"Volz",
"Kerstin",
""
],
[
"Reitzenstein",
"Stephan",
""
]
] |
2512.10074
|
Peyman Moeini
|
Peyman Moeini, Mark Coates
|
Cost and Complexity as Barriers to RTLS Adoption in SMEs: A Survey and Analysis
|
18 pages, 11 figures, 1 table. Survey of RTLS adoption barriers for small and medium-sized manufacturing enterprises
| null | null | null |
eess.SP
|
http://creativecommons.org/publicdomain/zero/1.0/
|
Real-time location systems (RTLSs) are central to Industry 4.0 and emerging Industry 5.0, providing the spatiotemporal data required for asset tracking, workflow optimization, safety, and integration with WMS, MES, and digital twins. While large enterprises increasingly deploy RTLSs, adoption among small and medium-sized enterprises (SMEs) remains limited. This paper examines whether cost and installation complexity are primary barriers to SME adoption. We position RTLSs within the broader Industry 4.0 and Logistics 4.0 landscape and summarize their operational value. We then synthesize evidence from the literature on technical, financial, and organizational constraints, with emphasis on infrastructure requirements, calibration effort, integration with legacy systems, and human factors. To complement this analysis, we report results from an online survey of sixteen manufacturing and technology professionals in Canada and the United States. Respondents report strong perceived value for real-time tracking but identify upfront cost, installation effort, integration difficulty, and reliance on multiple anchor nodes as dominant obstacles. Most indicate acceptable upfront investments below $10,000 and express a clear preference for low-infrastructure deployments with minimized anchor counts. Building on these findings, we outline design directions for SME-focused RTLSs, including wireless and modular architectures, cloud-managed and self-calibrating systems, standardized integration interfaces, and anchor-minimizing or anchor-free localization methods. Overall, the results show that limited SME adoption stems less from insufficient perceived value than from misalignment between current RTLS deployment models and SME resource constraints.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:51:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Moeini",
"Peyman",
""
],
[
"Coates",
"Mark",
""
]
] |
2512.10075
|
Jocelyn Nembe
|
Jocelyn Nemb\'e
|
Concentration of Measure under Diffeomorphism Groups: A Universal Framework with Optimal Coordinate Selection
| null | null | null | null |
math.ST stat.TH
|
http://creativecommons.org/licenses/by/4.0/
|
We establish a universal framework for concentration inequalities based on invariance under diffeomorphism groups. Given a probability measure $\mu$ on a space $E$ and a diffeomorphism $\psi: E \to F$, concentration properties transfer covariantly: if the pushforward $\psi_*\mu$ concentrates, so does $\mu$ in the pullback geometry. This reveals that classical concentration inequalities -- Hoeffding, Bernstein, Talagrand, Gaussian isoperimetry -- are manifestations of a single principle of \emph{geometric invariance}. The choice of coordinate system $\psi$ becomes a free parameter that can be optimized. We prove that for any distribution class $\Pc$, there exists an optimal diffeomorphism $\psi^*$ minimizing the concentration constant, and we characterize $\psi^*$ in terms of the Fisher-Rao geometry of $\Pc$. We establish \emph{strict improvement theorems}: for heavy-tailed or multiplicative data, the optimal $\psi$ yields exponentially tighter bounds than the identity. We develop the full theory including transportation-cost inequalities, isoperimetric profiles, and functional inequalities, all parametrized by the diffeomorphism group $\Diff(E)$. Connections to information geometry (Amari's $\alpha$-connections), optimal transport with general costs, and Riemannian concentration are established. Applications to robust statistics, multiplicative models, and high-dimensional inference demonstrate that coordinate optimization can improve statistical efficiency by orders of magnitude.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:54:05 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Nembé",
"Jocelyn",
""
]
] |
2512.10076
|
Pedro Picchetti
|
Luca Moreno-Louzada, Guilherme Figueira, Pedro Picchetti
|
Microfoundations and the Causal Interpretation of Price-Exposure Designs
| null | null | null | null |
econ.EM
|
http://creativecommons.org/licenses/by/4.0/
|
This paper studies regional exposure designs that use commodity prices as instruments to study local effects of aggregate shocks. Unlike standard shift share designs that leverage differential exposure to many shocks, the price exposure relies on exogenous variation from a single shock, leading to challenges for both identification and inference. We motivate the design using a multi sector labor model. Under the model and a potential outcomes framework, we characterize the 2SLS and TWFE estimands as weighted averages of region and sector specific effects plus contamination terms driven by the covariance structure of prices and by general-equilibrium output responses. We derive conditions under which these estimands have a clear causal interpretation and provide simple sensitivity analysis procedures for violations. Finally, we show that standard inference procedures suffer from an overrejection problem in price-exposure designs. We derive a new standard error estimator and show its desirable finite-sample properties through Monte Carlo simulations. In an application to gold mining and homicides in the Amazon, the price exposure standard errors are roughly twice as large as conventional clustered standard errors, making the main effect statistically insignificant.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:57:08 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Moreno-Louzada",
"Luca",
""
],
[
"Figueira",
"Guilherme",
""
],
[
"Picchetti",
"Pedro",
""
]
] |
2512.10077
|
Graham Denham
|
Graham Denham, Galen Dorpalen-Barry, Nicholas Proudfoot
|
Yoshinaga's criterion and the Varchenko--Gelfand algebra
|
12 pages, comments welcome
| null | null | null |
math.CO
|
http://creativecommons.org/licenses/by/4.0/
|
A central question in the theory of hyperplane arrangements is when the complement of a complex arrangement is aspherical. Yoshinaga provides a necessary condition in the case of complexifications of real arrangements, which he formulates purely combinatorially. We show that Yoshinaga's criterion is equivalent to a natural statement about the Varchenko-Gelfand ring, which in practice allows for fast calculation. We conclude with an investigation of the relationships between various properties of arrangements, including Yoshinaga's criterion and the asphericity of the arrangement complement.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 20:58:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Denham",
"Graham",
""
],
[
"Dorpalen-Barry",
"Galen",
""
],
[
"Proudfoot",
"Nicholas",
""
]
] |
2512.10078
|
Jingyao Ren
|
Jingyao Ren, Eric Ewing, T. K. Satish Kumar, Sven Koenig, Nora Ayanian
|
Empirical Hardness in Multi-Agent Pathfinding: Research Challenges and Opportunities
|
Published in AAMAS-25
|
Proceedings of the 24th International Conference on Autonomous Agents and Multiagent Systems, 2025, Pages 2885-2889
| null | null |
cs.MA
|
http://creativecommons.org/licenses/by/4.0/
|
Multi-agent pathfinding (MAPF) is the problem of finding collision-free paths for a team of agents on a map. Although MAPF is NP-hard, the hardness of solving individual instances varies significantly, revealing a gap between theoretical complexity and actual hardness. This paper outlines three key research challenges in MAPF empirical hardness to understand such phenomena. The first challenge, known as algorithm selection, is determining the best-performing algorithms for a given instance. The second challenge is understanding the key instance features that affect MAPF empirical hardness, such as structural properties like phase transition and backbone/backdoor. The third challenge is how to leverage our knowledge of MAPF empirical hardness to effectively generate hard MAPF instances or diverse benchmark datasets. This work establishes a foundation for future empirical hardness research and encourages deeper investigation into these promising and underexplored areas.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:02:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ren",
"Jingyao",
""
],
[
"Ewing",
"Eric",
""
],
[
"Kumar",
"T. K. Satish",
""
],
[
"Koenig",
"Sven",
""
],
[
"Ayanian",
"Nora",
""
]
] |
2512.10079
|
Federico Formica
|
Federico Formica, Mark Lawford, Claudio Menghi
|
Search-based Software Testing Driven by Domain Knowledge: Reflections and New Perspectives
| null | null | null | null |
cs.SE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Search-based Software Testing (SBST) can automatically generate test cases to search for requirements violations. Unlike manual test case development, it can generate a substantial number of test cases in a limited time. However, SBST does not possess the domain knowledge of engineers. Several techniques have been proposed to integrate engineers' domain knowledge within existing SBST frameworks. This paper will reflect on recent experimental results by highlighting bold and unexpected results. It will help re-examine SBST techniques driven by domain knowledge from a new perspective, suggesting new directions for future research.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:05:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Formica",
"Federico",
""
],
[
"Lawford",
"Mark",
""
],
[
"Menghi",
"Claudio",
""
]
] |
2512.10080
|
Luciano Floridi
|
Luciano Floridi, Jessica Morley, Claudio Novelli, David Watson
|
What Kind of Reasoning (if any) is an LLM actually doing? On the Stochastic Nature and Abductive Appearance of Large Language Models
| null | null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
This article looks at how reasoning works in current Large Language Models (LLMs) that function using the token-completion method. It examines their stochastic nature and their similarity to human abductive reasoning. The argument is that these LLMs create text based on learned patterns rather than performing actual abductive reasoning. When their output seems abductive, this is largely because they are trained on human-generated texts that include reasoning structures. Examples are used to show how LLMs can produce plausible ideas, mimic commonsense reasoning, and give explanatory answers without being grounded in truth, semantics, verification, or understanding, and without performing any real abductive reasoning. This dual nature, where the models have a stochastic base but appear abductive in use, has important consequences for how LLMs are evaluated and applied. They can assist with generating ideas and supporting human thinking, but their outputs must be critically assessed because they cannot identify truth or verify their explanations. The article concludes by addressing five objections to these points, noting some limitations in the analysis, and offering an overall evaluation.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:06:28 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Floridi",
"Luciano",
""
],
[
"Morley",
"Jessica",
""
],
[
"Novelli",
"Claudio",
""
],
[
"Watson",
"David",
""
]
] |
2512.10081
|
Kensuke Takii
|
Kensuke Takii, Changhao Liang, Hiroaki Ogata
|
Defining the Scope of Learning Analytics: An Axiomatic Approach for Analytic Practice and Measurable Learning Phenomena
|
27 pages, 1 figure, 2 tables
| null | null | null |
cs.CY cs.AI stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Learning Analytics (LA) has rapidly expanded through practical and technological innovation, yet its foundational identity has remained theoretically under-specified. This paper addresses this gap by proposing the first axiomatic theory that formally defines the essential structure, scope, and limitations of LA. Derived from the psychological definition of learning and the methodological requirements of LA, the framework consists of five axioms specifying discrete observation, experience construction, state transition, and inference. From these axioms, we derive a set of theorems and propositions that clarify the epistemological stance of LA, including the inherent unobservability of learner states, the irreducibility of temporal order, constraints on reachable states, and the impossibility of deterministically predicting future learning. We further define LA structure and LA practice as formal objects, demonstrating the sufficiency and necessity of the axioms and showing that diverse LA approaches -- such as Bayesian Knowledge Tracing and dashboards -- can be uniformly explained within this framework. The theory provides guiding principles for designing analytic methods and interpreting learning data while avoiding naive behaviorism and category errors by establishing an explicit theoretical inference layer between observations and states. This work positions LA as a rigorous science of state transition systems based on observability, establishing the theoretical foundation necessary for the field's maturation as a scholarly discipline.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:07:19 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Takii",
"Kensuke",
""
],
[
"Liang",
"Changhao",
""
],
[
"Ogata",
"Hiroaki",
""
]
] |
2512.10082
|
Federico Caucci
|
Federico Caucci
|
Kobayashi hyperbolicity of complete linear systems on abelian varieties
|
16 pages
| null | null | null |
math.AG math.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We provide a geometric condition ensuring that a very general element of a complete linear system on an abelian variety is Kobayashi hyperbolic. Some related conjectures are also given.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:08:51 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Caucci",
"Federico",
""
]
] |
2512.10083
|
Patrick Henning
|
Patrick Henning, Laura Huynh and Daniel Peterseim
|
Metric-driven numerical methods
| null | null | null | null |
math.NA cs.NA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, we explore the concept of metric-driven numerical methods as a powerful tool for solving various types of multiscale partial differential equations. Our focus is on computing constrained minimizers of functionals - or, equivalently, by considering the associated Euler-Lagrange equations - the solution of a class of eigenvalue problems that may involve nonlinearities in the eigenfunctions. We introduce metric-driven methods for such problems via Riemannian gradient techniques, leveraging the idea that gradients can be represented in different metrics (so-called Sobolev gradients) to accelerate convergence. We show that the choice of metric not only leads to specific metric-driven iterative schemes, but also induces approximation spaces with enhanced properties, particularly in low-regularity regimes or when the solution exhibits heterogeneous multiscale features. In fact, we recover a well-known class of multiscale spaces based on the Localized Orthogonal Decomposition (LOD), now derived from a new perspective. Alongside a discussion of the metric-driven approach for a model problem, we also demonstrate its application to simulating the ground states of spin-orbit-coupled Bose-Einstein condensates.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:10:37 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Henning",
"Patrick",
""
],
[
"Huynh",
"Laura",
""
],
[
"Peterseim",
"Daniel",
""
]
] |
2512.10084
|
Triparno Bandyopadhyay
|
Althaf M., Triparno Bandyopadhyay
|
Potential for the discovery of the protophobic boson at the STCF
|
20 pages, 19 figures
| null | null | null |
hep-ph hep-ex
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the morphology of the main drift chamber (MDC) proposed to be built around the collision point at the upcoming Super tau-charm facility (STCF), to check for its suitability in discovering the 17 MeV protophobic boson (X17 boson), hypothesised as a solution to the persistent ATOMKI nuclear-transition anomalies. These anomalies, observed in the excited $^8$Be, $^4$He, $^{12}$C, $^{16}$O nuclear transitions, have been interpreted as evidence for a $\sim$17 MeV, protophobic vector boson. Using the TrackEff framework, we perform detector-level simulations of the STCF MDC, and evaluate displaced-vertex sensitivities towards the protophobic boson, across the relevant mass-coupling parameter space. We study benchmark scenarios with visible and dark decay channels to perform likelihood-based significance estimates in order to determine the 5$\sigma$ discovery reach for the protophobic boson. We find that STCF can discover the protophobic boson while tolerating $\sim 10^4$ background events for specific regions of the parameter space around the 17 MeV peak. Our analysis establishes the first feasibility study of displaced light-boson searches at the STCF, motivating a full Geant-4 simulation.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:13:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"M.",
"Althaf",
""
],
[
"Bandyopadhyay",
"Triparno",
""
]
] |
2512.10085
|
Miguel Abadi
|
Miguel Abadi
|
Large Deviation inequalities for sums of positive correlated variables with clustering
|
21 pages
| null | null | null |
math.PR
|
http://creativecommons.org/licenses/by/4.0/
|
Large deviation inequalities for ergodic sums is an important subject since the seminal contribution of Bernstein for independent random variables with finite variances, followed by the Chernoff method and the Hoefding result for independent bounded variables. Very few results appears in the literature for the non independent case. Here we consider the, barely treated in the literature, case of positively correlated Bernoulli variables. This case represents the appearance in clusters of a certain fixed phenomena in the overlying stochastic process. Under a very mild condition we prove several upper deviation inequalities. The results follow by a spectral decomposition of an appropriated recursive operator. We illustrate with examples.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:16:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Abadi",
"Miguel",
""
]
] |
2512.10086
|
Huachen Sun
|
Yuri V. Kovchegov, M. Gabriel Santiago, Huachen Sun
|
Unpolarized GPDs at small $x$ and non-zero skewness
|
12 pages, 2 figures
| null | null | null |
hep-ph nucl-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the small-$x$ asymptotics of unpolarized generalized parton distributions (GPDs) and generalized transverse momentum distributions (GTMDs). Unlike the previous works in the literature, we consider the case of non-zero (but small) skewness while allowing for non-linear contributions to the evolution equations. We show that unpolarized GPDs and GTMDs at small $x$ are related to the eikonal dipole amplitude $N$, whose small-$x$ evolution is given by the BK/JIMWLK evolution equations, and to the odderon amplitude $\cal O$, whose evolution is also known in the literature. We show that the effect of non-zero skewness $\xi \neq 0$ is to modify the value of the evolution parameter (rapidity) in the arguments for the dipole amplitudes $N$ and $\cal O$ from $Y = \ln (1/x)$ to $Y = \ln \min \left\{ 1/|x| , 1/|\xi| \right\}$.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:16:58 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kovchegov",
"Yuri V.",
""
],
[
"Santiago",
"M. Gabriel",
""
],
[
"Sun",
"Huachen",
""
]
] |
2512.10087
|
Igor Rivin
|
Igor Rivin
|
Maximal Volume Ideal Polyhedra and the Arithmetic Angle Phenomenon
| null | null | null | null |
math.GT
|
http://creativecommons.org/licenses/by/4.0/
|
We present a software suite for the analysis and optimization of ideal convex polyhedra in hyperbolic 3-space $\mathbb{H}^3$. Using Rivin's variational characterization of ideal polyhedra, we develop efficient algorithms for checking combinatorial realizability and finding volume-maximizing configurations. Our systematic computational study reveals two striking phenomena: (1) maximal volume ideal polyhedra consistently exhibit dihedral angles that are rational multiples of $\pi$ -- a property with no obvious explanation from the optimization formulation; and (2) the distribution of volumes for random configurations is well-approximated by a Beta distribution, with mean normalized volume converging to approximately $\ln 2 \approx 0.69$ as the vertex count increases. We provide complete data for small vertex counts, including vertex positions, triangulations, and verified rational angle structures. An interactive implementation is publicly available.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:21:02 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Rivin",
"Igor",
""
]
] |
2512.10088
|
Anil Kumar Gorthi
|
Anil Kumar Gorthi
|
Evaluation of Risk and Resilience of the MBTA Green Rapid Transit System
|
28 pages, multiple figures. Course project (CY 5250, Northeastern University)
| null | null | null |
cs.CR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Transportation Systems Sector is one of the sixteen critical infrastructure sectors identified by the Cybersecurity and Infrastructure Security Agency (CISA) and plays a crucial role in ensuring public safety, economic stability, and national security. The Massachusetts Bay Transportation Authority (MBTA) serves as the primary public transportation system in the Greater Boston Area, with the Green Line representing one of the oldest and most complex rapid transit systems in the network. This paper presents a network-based risk and resilience assessment of the MBTA Green Line using graph theory, network metrics, and the Model-Based Risk Analysis (MBRA) tool. The original 70-station Green Line network is simplified into a 17-node model, and key metrics, including degree centrality, betweenness centrality, eigenvector centrality, spectral radius, node robustness, and blocking nodes, are computed using Python-based analysis. Critical vulnerability is derived using the MBRA resiliency equation, and random, targeted, and cyber-physical attack scenarios are evaluated. The results identify North Station, Government Center, Haymarket, Copley, and Kenmore as the most critical nodes. A fault tree analysis between Kenmore and Copley further demonstrates the impact of budget allocation on threat reduction. This work highlights key vulnerabilities in the Green Line network and provides actionable recommendations to improve resilience against cyber-physical threats.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:23:32 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gorthi",
"Anil Kumar",
""
]
] |
2512.10089
|
Jeongeun Kim
|
Jeongeun Kim, Sabrina Yarzada, Paul Chen, and Christopher Torng
|
Algorithm-Driven On-Chip Integration for High Density and Low Cost
| null | null | null | null |
cs.AR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Growing interest in semiconductor workforce development has generated demand for platforms capable of supporting large numbers of independent hardware designs for research and training without imposing high per-project overhead. Traditional multi-project wafer (MPW) services based solely on physical co-placement have historically met this need, yet their scalability breaks down as project counts rise. Recent efforts towards scalable chip tapeouts mitigate these limitations by integrating many small designs within a shared die and attempt to amortize costly resources such as IO pads and memory macros. However, foundational principles for arranging, linking, and validating such densely integrated design sites have received limited systematic investigation. This work presents a new approach with three key techniques to address this gap. First, we establish a structured formulation of the design space that enables automated, algorithm-driven packing of many projects, replacing manual layout practices. Second, we introduce an architecture that exploits only the narrow-area regions between sites to deliver on off-chip communication and other shared needs. Third, we provide a practical approach for on-chip power domains enabling per-project power characterization at a standard laboratory bench and requiring no expertise in low-power ASIC design. Experimental results show that our approach achieves substantial area reductions of up to 13x over state-of-the-art physical-only aggregation methods, offering a scalable and cost-effective path forward for large-scale tapeout environments.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:23:33 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kim",
"Jeongeun",
""
],
[
"Yarzada",
"Sabrina",
""
],
[
"Chen",
"Paul",
""
],
[
"Torng",
"Christopher",
""
]
] |
2512.10090
|
Dmytro Plutenko
|
Dmytro O. Plutenko and Mikhail V. Vasnetsov
|
Generation of Perfectly Achromatic Optical Vortices Using a Compensated Tandem Twisted Nematic Cell
|
20 pages, 5 figures
| null | null | null |
physics.optics cond-mat.soft
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The generation of white optical vortices is currently constrained by intrinsic trade-offs between spectral bandwidth, conversion efficiency, and temporal pulse integrity in conventional diffractive and geometric-phase approaches. In this work, we theoretically investigate a compensated tandem crossed twisted nematic (TN) liquid crystal architecture that overcomes these fundamental limitations. By developing a rigorous Jones matrix model and defining specific figures of merit for chromatic fidelity, we analyze the impact of manufacturing imperfections and non-adiabatic waveguiding (deviations from the Mauguin regime) on the device performance. We propose and evaluate three distinct compensation strategies, ranging from optimized passive designs for specific manufacturing tolerances to a robust active compensation scheme utilizing a tunable retarder. Our analysis demonstrates that the active approach effectively nullifies parasitic amplitude modulation, enabling the generation of perfectly achromatic vortices with high phase purity across an arbitrary bandwidth. This establishes the compensated tandem TN cell as a superior and versatile platform for high-fidelity white-light singular optics.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:23:37 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Plutenko",
"Dmytro O.",
""
],
[
"Vasnetsov",
"Mikhail V.",
""
]
] |
2512.10091
|
Silvia Pla Garc\'ia
|
Dimitrios Kranas, Amaury Marchon and Silvia Pla
|
Entanglement in the Schwinger effect
|
32 pages, 8 figures
| null | null | null |
hep-th gr-qc quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We analyze entanglement generated by the Schwinger effect using a mode-by-mode formalism for scalar and spinor QED in constant backgrounds. Starting from thermal initial states, we derive compact, closed-form results for bipartite entanglement between particle-antiparticle partners in terms of the Bogoliubov coefficients. For bosons, thermal fluctuations enhance production but suppress quantum correlations: the logarithmic negativity is nonzero only below a (mode-dependent) critical temperature $T_c$. At fixed $T$, entanglement appears only above a critical field $E_{\text{crit,entang}}$. For fermions, we observe a qualitatively different pattern: at finite $T$ entanglement exists only within a finite window $E_{\text{min}} < E < E_{\text{max}}$, with a temperature-independent optimal field strength $E_{*}$ that maximizes the logarithmic negativity. Entanglement is vanishing above $T_{\text{max}}=\omega/\text{arcsinh}(1)$. We give quantitative estimates for analog experiments, where our entanglement criteria convert directly into concrete temperature and electric field constraints. These findings identify realistic regimes where the quantum character of Schwinger physics may be tested in the laboratory.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:24:02 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kranas",
"Dimitrios",
""
],
[
"Marchon",
"Amaury",
""
],
[
"Pla",
"Silvia",
""
]
] |
2512.10092
|
Nick Jiang
|
Nick Jiang, Xiaoqing Sun, Lisa Dunlap, Lewis Smith, Neel Nanda
|
Interpretable Embeddings with Sparse Autoencoders: A Data Analysis Toolkit
|
Code: https://github.com/nickjiang2378/interp_embed
| null | null | null |
cs.AI cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Analyzing large-scale text corpora is a core challenge in machine learning, crucial for tasks like identifying undesirable model behaviors or biases in training data. Current methods often rely on costly LLM-based techniques (e.g. annotating dataset differences) or dense embedding models (e.g. for clustering), which lack control over the properties of interest. We propose using sparse autoencoders (SAEs) to create SAE embeddings: representations whose dimensions map to interpretable concepts. Through four data analysis tasks, we show that SAE embeddings are more cost-effective and reliable than LLMs and more controllable than dense embeddings. Using the large hypothesis space of SAEs, we can uncover insights such as (1) semantic differences between datasets and (2) unexpected concept correlations in documents. For instance, by comparing model responses, we find that Grok-4 clarifies ambiguities more often than nine other frontier models. Relative to LLMs, SAE embeddings uncover bigger differences at 2-8x lower cost and identify biases more reliably. Additionally, SAE embeddings are controllable: by filtering concepts, we can (3) cluster documents along axes of interest and (4) outperform dense embeddings on property-based retrieval. Using SAE embeddings, we study model behavior with two case studies: investigating how OpenAI model behavior has changed over time and finding "trigger" phrases learned by Tulu-3 (Lambert et al., 2024) from its training data. These results position SAEs as a versatile tool for unstructured data analysis and highlight the neglected importance of interpreting models through their data.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:26:24 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Jiang",
"Nick",
""
],
[
"Sun",
"Xiaoqing",
""
],
[
"Dunlap",
"Lisa",
""
],
[
"Smith",
"Lewis",
""
],
[
"Nanda",
"Neel",
""
]
] |
2512.10093
|
Oleg Morzhin
|
Oleg V. Morzhin
|
Gradient projection method and stochastic search for some optimal control models with spin chains. I
|
15 pages, 2 figures
| null | null | null |
quant-ph math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This article (I) considers the known optimal control model of a quantum information transfer along a spin chain with controlled external parabolic magnetic field, with an arbitrary length. The article adds certain lower and upper pointwise constraints on controls, adds the problem of keeping the signal at the last spin, considers various classes of controls. For these problems under piecewise continuous controls, the projection-type linearized Pontryagin maximum principle, gradient projection method's constructions in its one- and two- and three-step forms were adapted by analogy with [Morzhin O.V., Pechen A.N. J. Phys. A: Math. Theor., 2025]. Moreover, an example with a genetic algorithm's successful use under a special class of controls is given.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:32:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Morzhin",
"Oleg V.",
""
]
] |
2512.10094
|
Brian Zhu
|
Brian Zhu
|
Does Timeboost Reduce MEV-Related Spam? Theory and Evidence from Layer-2 Transactions
| null | null | null | null |
cs.GT
|
http://creativecommons.org/licenses/by/4.0/
|
Maximal extractable value opportunities often induce spam in Layer-2 blockchains: many identical transactions are submitted near simultaneously, most of which revert, wasting blockspace. We study Timeboost, a mechanism on Arbitrum that auctions a timestamp advantage, crucial under first-come first-served sequencing rules. We develop a game-theoretic model in which users choose the number of transaction copies to submit, and extend upon the baseline setting by modeling the Timeboost auction and subsequent transaction submission behavior. We show that Timeboost reduces spam and increases sequencer/DAO revenue in equilibrium relative to the baseline, transferring user payments from revert costs to auction bids. Empirically, we assemble mempool data from multiple Layer-2 networks, measuring spam via identical transactions submitted in narrow time intervals, and conduct an event study around Timeboost adoption on Arbitrum using other L2s as contemporaneous benchmarks. We find a decline in MEV-related spam and an increase in revenue on Arbitrum post-adoption, consistent with model predictions.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:33:07 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhu",
"Brian",
""
]
] |
2512.10095
|
Jiachen Tao
|
Jiachen Tao, Junyi Wu, Haoxuan Wang, Zongxin Yang, Dawen Cai, Yan Yan
|
TraceFlow: Dynamic 3D Reconstruction of Specular Scenes Driven by Ray Tracing
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
We present TraceFlow, a novel framework for high-fidelity rendering of dynamic specular scenes by addressing two key challenges: precise reflection direction estimation and physically accurate reflection modeling. To achieve this, we propose a Residual Material-Augmented 2D Gaussian Splatting representation that models dynamic geometry and material properties, allowing accurate reflection ray computation. Furthermore, we introduce a Dynamic Environment Gaussian and a hybrid rendering pipeline that decomposes rendering into diffuse and specular components, enabling physically grounded specular synthesis via rasterization and ray tracing. Finally, we devise a coarse-to-fine training strategy to improve optimization stability and promote physically meaningful decomposition. Extensive experiments on dynamic scene benchmarks demonstrate that TraceFlow outperforms prior methods both quantitatively and qualitatively, producing sharper and more realistic specular reflections in complex dynamic environments.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:36:32 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tao",
"Jiachen",
""
],
[
"Wu",
"Junyi",
""
],
[
"Wang",
"Haoxuan",
""
],
[
"Yang",
"Zongxin",
""
],
[
"Cai",
"Dawen",
""
],
[
"Yan",
"Yan",
""
]
] |
2512.10097
|
Karri Folan Di Petrillo
|
Mira Littmann, Mark Larson, Benjamin Rosser, Tate Flicker, Kane Huang, Leo Rozanov, and Karri Folan Di Petrillo
|
Enabling searches for long-lived particles at a future 10 TeV Muon Collider
| null | null | null | null |
hep-ex physics.ins-det
|
http://creativecommons.org/licenses/by/4.0/
|
Muon Colliders offer fantastic opportunities to explore new phenomena at the energy frontier. However, beam-induced-backgrounds from muon decays pose significant challenges for detector design, readout, and reconstruction. Previous detector studies have employed stringent hit-timing requirements to reduce occupancy to manageable levels with negligible efficiency loss for prompt Standard Model particles. In the spirit of maximizing discovery potential, we investigate the capability of detecting meta-stable charged long-lived particles at a 10 TeV Muon Collider. As a benchmark, we consider a Gauge Mediated Supersymmetry Breaking (GMSB) model in which the stau is long-lived and can be identified as a high momentum, slowly moving track. We find that nominal hit-timing selections are too restrictive, and investigate the impact of looser requirements. We demonstrate that it is possible to recover sensitivity to particles with masses close to $\sqrt{s}/2$ by expanding time acceptance, and provide recommendations to further improve tracker design and track reconstruction.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:39:20 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Littmann",
"Mira",
""
],
[
"Larson",
"Mark",
""
],
[
"Rosser",
"Benjamin",
""
],
[
"Flicker",
"Tate",
""
],
[
"Huang",
"Kane",
""
],
[
"Rozanov",
"Leo",
""
],
[
"Di Petrillo",
"Karri Folan",
""
]
] |
2512.10099
|
Steven Caro
|
Steven Caro, Stephen L. Smith
|
Push Smarter, Not Harder: Hierarchical RL-Diffusion Policy for Efficient Nonprehensile Manipulation
|
8 pages, 8 figures
| null | null | null |
cs.RO cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Nonprehensile manipulation, such as pushing objects across cluttered environments, presents a challenging control problem due to complex contact dynamics and long-horizon planning requirements. In this work, we propose HeRD, a hierarchical reinforcement learning-diffusion policy that decomposes pushing tasks into two levels: high-level goal selection and low-level trajectory generation. We employ a high-level reinforcement learning (RL) agent to select intermediate spatial goals, and a low-level goal-conditioned diffusion model to generate feasible, efficient trajectories to reach them.
This architecture combines the long-term reward maximizing behaviour of RL with the generative capabilities of diffusion models. We evaluate our method in a 2D simulation environment and show that it outperforms the state-of-the-art baseline in success rate, path efficiency, and generalization across multiple environment configurations. Our results suggest that hierarchical control with generative low-level planning is a promising direction for scalable, goal-directed nonprehensile manipulation. Code, documentation, and trained models are available: https://github.com/carosteven/HeRD.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:40:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Caro",
"Steven",
""
],
[
"Smith",
"Stephen L.",
""
]
] |
2512.10100
|
Apostol Vassilev
|
Apostol Vassilev
|
Robust AI Security and Alignment: A Sisyphean Endeavor?
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
This manuscript establishes information-theoretic limitations for robustness of AI security and alignment by extending G\"odel's incompleteness theorem to AI. Knowing these limitations and preparing for the challenges they bring is critically important for the responsible adoption of the AI technology. Practical approaches to dealing with these challenges are provided as well. Broader implications for cognitive reasoning limitations of AI systems are also proven.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:44:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Vassilev",
"Apostol",
""
]
] |
2512.10101
|
Koen Schouten
|
Koen Schouten and Mikhail Isachenkov
|
The von Neumann algebraic quantum group $\mathrm{SU}_q(1,1)\rtimes \mathbb{Z}_2$ and the DSSYK model
|
89 pages, 4 figures
| null | null | null |
math-ph cond-mat.stat-mech hep-th math.MP math.OA math.QA
|
http://creativecommons.org/licenses/by/4.0/
|
The double-scaling limit of the SYK (DSSYK) model is known to possess an underlying $\mathcal{U}_q(\mathfrak{su}(1,1))$ quantum group symmetry. In this paper, we provide, for the first time, a von Neumann algebraic quantum group-theoretical description of the degrees of freedom and the dynamics of the DSSYK model. In particular, we construct the operator-algebraic quantum Gauss decomposition for the von Neumann algebraic quantum group $\mathrm{SU}_q(1,1)\rtimes \mathbb{Z}_2$, i.e. the $q$-deformation of the normaliser of $\mathrm{SU}(1,1)$ in $\mathrm{SL}(2,\mathbb{C})$, and derive the Casimir action on its quantum homogeneous spaces. We then show that the dynamics on quantum AdS$_{2,q}$ space reduces to that of the DSSYK model. Furthermore, we argue that the extension of the global symmetry group to its normaliser is not only necessary for a consistent definition of the locally compact quantum group, but that, moreover, the reduction to the DSSYK model works exclusively at the level of the normaliser. The von Neumann algebraic description is shown to give a natural restriction on the allowed quantised coordinates, elegantly ensuring length positivity and non-negative integer chord numbers. Lastly, we make remarks on the correlation function related to the strange series representation, which is argued to interpolate between the AdS and dS regions of our $q$-homogeneous space.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:46:07 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Schouten",
"Koen",
""
],
[
"Isachenkov",
"Mikhail",
""
]
] |
2512.10102
|
Neelima Prasad
|
Neelima Prasad, Jarek Reynolds, Neel Karsanbhai, Tanusree Sharma, Lotus Zhang, Abigale Stangl, Yang Wang, Leah Findlater, Danna Gurari
|
Hierarchical Instance Tracking to Balance Privacy Preservation with Accessible Information
|
Accepted at WACV 2026
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
We propose a novel task, hierarchical instance tracking, which entails tracking all instances of predefined categories of objects and parts, while maintaining their hierarchical relationships. We introduce the first benchmark dataset supporting this task, consisting of 2,765 unique entities that are tracked in 552 videos and belong to 40 categories (across objects and parts). Evaluation of seven variants of four models tailored to our novel task reveals the new dataset is challenging. Our dataset is available at https://vizwiz.org/tasks-and-datasets/hierarchical-instance-tracking/
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:48:04 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Prasad",
"Neelima",
""
],
[
"Reynolds",
"Jarek",
""
],
[
"Karsanbhai",
"Neel",
""
],
[
"Sharma",
"Tanusree",
""
],
[
"Zhang",
"Lotus",
""
],
[
"Stangl",
"Abigale",
""
],
[
"Wang",
"Yang",
""
],
[
"Findlater",
"Leah",
""
],
[
"Gurari",
"Danna",
""
]
] |
2512.10103
|
Jiayin Zhang
|
Jiayin Zhang, Anthony J. Ardizzi, Kieran A. Cleary, Austin J. Minnich
|
Investigation of Real-Space Transfer Noise in InP Quantum Wells
|
20 pages, 5 figures, 7 equations
| null | null | null |
physics.app-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Indium phosphide (InP) high electron-mobility transistors (HEMTs) are widely used in many fields such as quantum computing because of their unparalleled microwave noise performance. Achieving improved noise performance requires a physical understanding of the noise mechanisms. Here, we experimentally test a theoretical proposal for drain (output) noise as originating in part from real-space transfer (RST) by characterizing the microwave noise temperature of transfer-length method structures with the same channel composition but two different barrier compositions. This choice was made to alter the confining potential of electrons in the channel, thereby affecting the RST mechanism, while avoiding changes to the channel transport properties. We observe trends of noise temperature with physical temperature and source-drain voltage which are compatible with the predictions of RST noise theory. This finding supports the hypothesis that RST contributes to drain noise in HEMTs.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:49:08 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhang",
"Jiayin",
""
],
[
"Ardizzi",
"Anthony J.",
""
],
[
"Cleary",
"Kieran A.",
""
],
[
"Minnich",
"Austin J.",
""
]
] |
2512.10104
|
Shaohu Zhang
|
Najmul Hassan, Prashanth BusiReddyGari, Haitao Zhao, Yihao Ren, Jinsheng Xu, Shaohu Zhang
|
LLM-PEA: Leveraging Large Language Models Against Phishing Email Attacks
|
7 pages
| null | null | null |
cs.CR cs.IR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Email phishing is one of the most prevalent and globally consequential vectors of cyber intrusion. As systems increasingly deploy Large Language Models (LLMs) applications, these systems face evolving phishing email threats that exploit their fundamental architectures. Current LLMs require substantial hardening before deployment in email security systems, particularly against coordinated multi-vector attacks that exploit architectural vulnerabilities. This paper proposes LLMPEA, an LLM-based framework to detect phishing email attacks across multiple attack vectors, including prompt injection, text refinement, and multilingual attacks. We evaluate three frontier LLMs (e.g., GPT-4o, Claude Sonnet 4, and Grok-3) and comprehensive prompting design to assess their feasibility, robustness, and limitations against phishing email attacks. Our empirical analysis reveals that LLMs can detect the phishing email over 90% accuracy while we also highlight that LLM-based phishing email detection systems could be exploited by adversarial attack, prompt injection, and multilingual attacks. Our findings provide critical insights for LLM-based phishing detection in real-world settings where attackers exploit multiple vulnerabilities in combination.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:50:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Hassan",
"Najmul",
""
],
[
"BusiReddyGari",
"Prashanth",
""
],
[
"Zhao",
"Haitao",
""
],
[
"Ren",
"Yihao",
""
],
[
"Xu",
"Jinsheng",
""
],
[
"Zhang",
"Shaohu",
""
]
] |
2512.10105
|
Navin Kumar
|
Soorya Ram Shimgekar, Abhay Goyal, Lam Yin Cheung, Roy Ka-Wei Lee, Koustuv Saha, Pi Zonooz, Navin Kumar
|
Modeling Narrative Archetypes in Conspiratorial Narratives: Insights from Singapore-Based Telegram Groups
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Conspiratorial discourse is increasingly embedded within digital communication ecosystems, yet its structure and spread remain difficult to study. This work analyzes conspiratorial narratives in Singapore-based Telegram groups, showing that such content is woven into everyday discussions rather than confined to isolated echo chambers. We propose a two-stage computational framework. First, we fine-tune RoBERTa-large to classify messages as conspiratorial or not, achieving an F1-score of 0.866 on 2,000 expert-labeled messages. Second, we build a signed belief graph in which nodes represent messages and edge signs reflect alignment in belief labels, weighted by textual similarity. We introduce a Signed Belief Graph Neural Network (SiBeGNN) that uses a Sign Disentanglement Loss to learn embeddings that separate ideological alignment from stylistic features.
Using hierarchical clustering on these embeddings, we identify seven narrative archetypes across 553,648 messages: legal topics, medical concerns, media discussions, finance, contradictions in authority, group moderation, and general chat. SiBeGNN yields stronger clustering quality (cDBI = 8.38) than baseline methods (13.60 to 67.27), supported by 88 percent inter-rater agreement in expert evaluations. Our analysis shows that conspiratorial messages appear not only in clusters focused on skepticism or distrust, but also within routine discussions of finance, law, and everyday matters. These findings challenge common assumptions about online radicalization by demonstrating that conspiratorial discourse operates within ordinary social interaction. The proposed framework advances computational methods for belief-driven discourse analysis and offers applications for stance detection, political communication studies, and content moderation policy.
|
[
{
"version": "v1",
"created": "Wed, 10 Dec 2025 21:51:16 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Shimgekar",
"Soorya Ram",
""
],
[
"Goyal",
"Abhay",
""
],
[
"Cheung",
"Lam Yin",
""
],
[
"Lee",
"Roy Ka-Wei",
""
],
[
"Saha",
"Koustuv",
""
],
[
"Zonooz",
"Pi",
""
],
[
"Kumar",
"Navin",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.