English

EvalCards: A Framework for Standardized Evaluation Reporting

Computation and Language 2025-12-01 v1 Artificial Intelligence Computers and Society

Abstract

Evaluation has long been a central concern in NLP, and transparent reporting practices are more critical than ever in today's landscape of rapidly released open-access models. Drawing on a survey of recent work on evaluation and documentation, we identify three persistent shortcomings in current reporting practices: reproducibility, accessibility, and governance. We argue that existing standardization efforts remain insufficient and introduce Evaluation Disclosure Cards (EvalCards) as a path forward. EvalCards are designed to enhance transparency for both researchers and practitioners while providing a practical foundation to meet emerging governance requirements.

Keywords

Cite

@article{arxiv.2511.21695,
  title  = {EvalCards: A Framework for Standardized Evaluation Reporting},
  author = {Ruchira Dhar and Danae Sanchez Villegas and Antonia Karamolegkou and Alice Schiavone and Yifei Yuan and Xinyi Chen and Jiaang Li and Stella Frank and Laura De Grazia and Monorama Swain and Stephanie Brandl and Daniel Hershcovich and Anders Søgaard and Desmond Elliott},
  journal= {arXiv preprint arXiv:2511.21695},
  year   = {2025}
}

Comments

Under review

R2 v1 2026-07-01T07:56:47.124Z