DFM Mimir v1: An Open HRM Delivering Frontier Performance at 1B Parameters Using Only Permissible Post-Training Data Paper • 2608.13517 • Published 21 days ago • 32
Improving reasoning at inference time via uncertainty minimisation Paper • 2603.07159 • Published Mar 7 • 1
Maintaining MTEB: Towards Long Term Usability and Reproducibility of Embedding Benchmarks Paper • 2506.21182 • Published Jun 26, 2025 • 3
One prompt is not enough: Instruction Sensitivity Undermines Embedding Model Evaluation Paper • 2605.22544 • Published May 21 • 2
One prompt is not enough: Instruction Sensitivity Undermines Embedding Model Evaluation Paper • 2605.22544 • Published May 21 • 2
MMTEB: Massive Multilingual Text Embedding Benchmark Paper • 2502.13595 • Published Feb 19, 2025 • 50
Dynaword: From One-shot to Continuously Developed Datasets Paper • 2508.02271 • Published Aug 4, 2025 • 15