Saved in:
Bibliographic Details
Main Authors: Li, Zongxia, Mondal, Ishani, Liang, Yijun, Nghiem, Huy, Boyd-Graber, Jordan Lee
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.11161
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909346232795136
author Li, Zongxia
Mondal, Ishani
Liang, Yijun
Nghiem, Huy
Boyd-Graber, Jordan Lee
author_facet Li, Zongxia
Mondal, Ishani
Liang, Yijun
Nghiem, Huy
Boyd-Graber, Jordan Lee
contents 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).
format Preprint
id arxiv_https___arxiv_org_abs_2402_11161
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle PEDANTS: Cheap but Effective and Interpretable Answer Equivalence
Li, Zongxia
Mondal, Ishani
Liang, Yijun
Nghiem, Huy
Boyd-Graber, Jordan Lee
Computation and Language
Artificial Intelligence
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).
title PEDANTS: Cheap but Effective and Interpretable Answer Equivalence
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2402.11161