PEDANTS: Cheap but Effective and Interpretable Answer Equivalence
Zongxia Li, Ishani Mondal, Huy Nghiem, Yijun Liang, Jordan Lee Boyd-Graber
Abstract
Question answering (QA) can only make progress if we know if an answer is correct, but current answer correctness (AC) metrics struggle with verbose, free-form answers from large language models (LLMs). There are two challenges with current short-form QA evaluations: a lack of diverse styles of evaluation data and an over-reliance on expensive and slow LLMs. LLM-based scorers correlate better with humans, but this expensive task has only been tested on limited QA datasets. We rectify these issues by providing rubrics and datasets for evaluating machine QA adopted from the Trivia community. We also propose an efficient, and interpretable QA evaluation that is more stable than an exact match and neural methods (BERTScore).- Anthology ID:
- 2024.findings-emnlp.548
- Volume:
- Findings of the Association for Computational Linguistics: EMNLP 2024
- Month:
- November
- Year:
- 2024
- Address:
- Miami, Florida, USA
- Editors:
- Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 9373–9398
- Language:
- URL:
- https://preview.aclanthology.org/remove-affiliations/2024.findings-emnlp.548/
- DOI:
- 10.18653/v1/2024.findings-emnlp.548
- Cite (ACL):
- Zongxia Li, Ishani Mondal, Huy Nghiem, Yijun Liang, and Jordan Lee Boyd-Graber. 2024. PEDANTS: Cheap but Effective and Interpretable Answer Equivalence. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 9373–9398, Miami, Florida, USA. Association for Computational Linguistics.
- Cite (Informal):
- PEDANTS: Cheap but Effective and Interpretable Answer Equivalence (Li et al., Findings 2024)
- PDF:
- https://preview.aclanthology.org/remove-affiliations/2024.findings-emnlp.548.pdf