PreprintCC BY 4.0Zenodo
Contract clause extraction: CUAD (Contract Understanding Atticus Dataset) 510: Regex Wins
Abstract
CUAD clause extraction benchmark compares Transformer and regex F1 scores, revealing when rule-based span labeling still beats neural models.
Independent technical note mirroring the canonical version: https://legalpdf.io/blog/contract-clause-extraction-cuad-contract-understanding-atticus-dataset-510-regex-wins.php
Related
- OSC700: A GW+BSE Dataset of 700 Organic Semiconductor (OSC) Crystals with an LLM-Assisted Data Extraction Agent
- A benchmark dataset for powershell fileless malware detection: methodology, validation, and n-Gram feature extraction
- A Canadian Benchmark LiDAR Dataset for Urban Infrastructure and 3D Scene Understanding
- Active Visual Semantics: A large-scale MEG and eye-tracking dataset for understanding visual intelligence in action
- Indemnification Clause Review 2026: Policy Guide for 90% vs 95% Auto-Accept Threshold
BibTeX
@misc{howard2026contract,
title = {{Contract clause extraction: CUAD (Contract Understanding Atticus Dataset) 510: Regex Wins}},
author = {Justin Howard},
howpublished = {Zenodo},
year = {2026},
month = oct,
doi = {10.5281/zenodo.23676389},
url = {https://zenodo.org/records/23676389},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Contract clause extraction: CUAD (Contract Understanding Atticus Dataset) 510: Regex Wins
AU - Justin Howard
PY - 2026
DA - 2026/10/26
PB - Zenodo
DO - 10.5281/zenodo.23676389
UR - https://zenodo.org/records/23676389
AB - CUAD clause extraction benchmark compares Transformer and regex F1 scores, revealing when rule-based span labeling still beats neural models. Independent technical note mirroring the canonical version: https://legalpdf.io/blog/contract-clause-extraction-cuad-contract-understanding-atticus-dataset-510-regex-wins.php
ER -