DLT-Scientific-Literature

Paper | GitHub

Dataset Description

Dataset Summary

DLT-Scientific-Literature is a specialized corpus of academic publications focused on Distributed Ledger Technology (DLT). This dataset is part of the larger DLT-Corpus collection, designed to support NLP research, language model development, and innovation studies in the DLT domain.

The dataset contains 37,440 scientific documents with 564 million tokens, spanning publications from 1978 to 2025. All documents are in English and have been filtered for domain relevance using a fine-tuned BERT model.

This dataset is part of the DLT-Corpus collection. For the complete corpus including patents and social media data, see: https://huggingface.co/collections/ExponentialScience/dlt-corpus-68e44e40d4e7a3bd7a224402

Languages

English (en)

Dataset Structure

Data Fields

The dataset includes the following fields for each document:

• paperId: Unique identifier from Semantic Scholar
• title: Title of the scientific publication
• authors: List of authors
• year: Publication year
• publicationDate: Full publication date
• venue: Publication venue (journal, conference, etc.)
• publicationVenue: Detailed venue information
• publicationTypes: Type of publication (e.g., JournalArticle, Conference)
• abstract: Abstract of the publication
• text: Full text content in Markdown format
• url: URL to the source document
• openAccessPdf: Link to open access PDF if available
• isOpenAccess: Boolean indicating open access status
• fieldsOfStudy: Academic fields associated with the paper
• s2FieldsOfStudy: Semantic Scholar's field classifications
• references: List of referenced papers
• lang: Language code
• lang_conf: Confidence score for language detection
• tok_len: Token length of the document
• total_tokens: Total number of tokens

Data Splits

This is a single corpus without predefined splits. Users should create their own train/validation/test splits based on their specific research needs.

Dataset Creation

Curation Rationale

DLT-Scientific-Literature was created to address the lack of large-scale, domain-specific text corpora for NLP and computational research in the Distributed Ledger Technology field. The dataset enables researchers to:

• Develop DLT-specific language models and embeddings
• Conduct innovation studies and trend analysis
• Perform text mining on cutting-edge DLT research
• Study the evolution of concepts and terminology in the field

Source Data

Data Collection

Scientific literature was collected from the Semantic Scholar API using domain-specific queries related to blockchain, distributed ledgers, cryptocurrencies, smart contracts, and related technologies.

Data Processing

The collection process involved:

1. Query-based retrieval: Using targeted keywords to retrieve relevant publications
2. PDF parsing: Converting PDF documents to Markdown format
3. Language detection: Filtering for English-language documents
4. Length filtering: Removing documents that are too short or too long
5. Domain relevance filtering: Using a fine-tuned BERT model to ensure documents are relevant to DLT

Personal and Sensitive Information

This dataset contains only publicly available scientific literature. No personal or confidential data is included. Author names and affiliations are retained as they appear in the original publications, as this is standard academic practice.

Considerations for Using the Data

Discussion of Biases

Potential biases include:

• Geographic bias: Publications may be skewed toward institutions in certain countries
• Language bias: Only English-language publications are included
• Temporal bias: More recent years may have disproportionately more publications
• Venue bias: Certain journals or conferences may be over-represented
• Citation bias: Highly-cited papers may be more likely to be included

Other Known Limitations

• Temporal coverage: While the dataset spans 1978-2025, the distribution is uneven with more recent years heavily represented
• Access limitations: Some publications may be missing due to access restrictions or API limitations
• Quality variation: Academic writing quality and rigor vary across publications
• Parsing errors: PDF-to-Markdown conversion may introduce formatting issues in some documents

Additional Information

Dataset Curators

Walter Hernandez Cruz, Peter Devine, Nikhil Vadgama, Paolo Tasca, Jiahua Xu

Licensing Information

Mixed open-access licenses including:

• Creative Commons Attribution (CC-BY)
• Creative Commons Attribution-ShareAlike (CC-BY-SA)
• Creative Commons Zero (CC0)
• Other permissive open-access licenses

Individual license information is included in the metadata for each document where available. Users should check the specific license for each document before use.

Citation Information

@misc{hernandez2026dlt-corpus,
      title={DLT-Corpus: A Large-Scale Text Collection for the Distributed Ledger Technology Domain}, 
      author={Walter Hernandez Cruz and Peter Devine and Nikhil Vadgama and Paolo Tasca and Jiahua Xu},
      year={2026},
      eprint={2602.22045},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2602.22045}, 
}