An Empirical Study of Evaluating Long-form Question Answering

• URL: http://arxiv.org/abs/2504.18413v1
• Date: Fri, 25 Apr 2025 15:14:25 GMT
• Title: An Empirical Study of Evaluating Long-form Question Answering
• Authors: Ning Xian, Yixing Fan, Ruqing Zhang, Maarten de Rijke, Jiafeng Guo,
• Abstract summary: We collect 5,236 factoid and non-factoid long-form answers generated by different large language models.<n>We conduct a human evaluation on 2,079 of them, focusing on correctness and informativeness.<n>We find that the style, length of the answers, and the category of questions can bias the automatic evaluation metrics.
• Score: 77.8023489322551
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: \Ac{LFQA} aims to generate lengthy answers to complex questions. This scenario presents great flexibility as well as significant challenges for evaluation. Most evaluations rely on deterministic metrics that depend on string or n-gram matching, while the reliability of large language model-based evaluations for long-form answers remains relatively unexplored. We address this gap by conducting an in-depth study of long-form answer evaluation with the following research questions: (i) To what extent do existing automatic evaluation metrics serve as a substitute for human evaluations? (ii) What are the limitations of existing evaluation metrics compared to human evaluations? (iii) How can the effectiveness and robustness of existing evaluation methods be improved? We collect 5,236 factoid and non-factoid long-form answers generated by different large language models and conduct a human evaluation on 2,079 of them, focusing on correctness and informativeness. Subsequently, we investigated the performance of automatic evaluation metrics by evaluating these answers, analyzing the consistency between these metrics and human evaluations. We find that the style, length of the answers, and the category of questions can bias the automatic evaluation metrics. However, fine-grained evaluation helps mitigate this issue on some metrics. Our findings have important implications for the use of large language models for evaluating long-form question answering. All code and datasets are available at https://github.com/bugtig6351/lfqa_evaluation.

Related papers

• Accurate and Nuanced Open-QA Evaluation Through Textual Entailment [4.762213968673381]
  We propose to study the entailment relations of answers to identify more informative and more general system answers. The entailment-based evaluation we propose allows the assignment of bonus or partial marks by quantifying the inference gap between answers.
  arXiv  Detail & Related papers  (2024-05-26T21:33:27Z)
• Is Reference Necessary in the Evaluation of NLG Systems? When and Where? [58.52957222172377]
  We show that reference-free metrics exhibit a higher correlation with human judgment and greater sensitivity to deficiencies in language quality. Our study can provide insight into the appropriate application of automatic metrics and the impact of metric choice on evaluation performance.
  arXiv  Detail & Related papers  (2024-03-21T10:31:11Z)
• Learning and Evaluating Human Preferences for Conversational Head Generation [101.89332968344102]
  We propose a novel learning-based evaluation metric named Preference Score (PS) for fitting human preference according to the quantitative evaluations across different dimensions. PS can serve as a quantitative evaluation without the need for human annotation.
  arXiv  Detail & Related papers  (2023-07-20T07:04:16Z)
• Style Over Substance: Evaluation Biases for Large Language Models [17.13064447978519]
  This study investigates the behavior of crowd-sourced and expert annotators, as well as large language models (LLMs) Our findings reveal a concerning bias in the evaluation process, as answers with factual errors are rated more favorably than answers that are too short or contained grammatical errors. We propose independently evaluating machine-generated text across multiple dimensions, rather than merging all the evaluation aspects into a single score.
  arXiv  Detail & Related papers  (2023-07-06T14:42:01Z)
• A Critical Evaluation of Evaluations for Long-form Question Answering [48.51361567469683]
  Long-form question answering (LFQA) enables answering a wide range of questions, but its flexibility poses enormous challenges for evaluation. We perform the first targeted study of the evaluation of long-form answers, covering both human and automatic evaluation practices.
  arXiv  Detail & Related papers  (2023-05-29T16:54:24Z)
• DEnsity: Open-domain Dialogue Evaluation Metric using Density Estimation [24.224114300690758]
  We propose DEnsity, which evaluates a response by utilizing density estimation on the feature space derived from a neural classifier. Our metric measures how likely a response would appear in the distribution of human conversations.
  arXiv  Detail & Related papers  (2023-05-08T14:10:40Z)
• Just Rank: Rethinking Evaluation with Word and Sentence Similarities [105.5541653811528]
  intrinsic evaluation for embeddings lags far behind, and there has been no significant update since the past decade. This paper first points out the problems using semantic similarity as the gold standard for word and sentence embedding evaluations. We propose a new intrinsic evaluation method called EvalRank, which shows a much stronger correlation with downstream tasks.
  arXiv  Detail & Related papers  (2022-03-05T08:40:05Z)
• Human Evaluation of Creative NLG Systems: An Interdisciplinary Survey on Recent Papers [0.685316573653194]
  We survey human evaluation in papers presenting work on creative natural language generation. The most typical human evaluation method is a scaled survey, typically on a 5 point scale. The most commonly evaluated parameters are meaning, syntactic correctness, novelty, relevance and emotional value.
  arXiv  Detail & Related papers  (2021-07-31T18:54:30Z)
• GO FIGURE: A Meta Evaluation of Factuality in Summarization [131.1087461486504]
  We introduce GO FIGURE, a meta-evaluation framework for evaluating factuality evaluation metrics. Our benchmark analysis on ten factuality metrics reveals that our framework provides a robust and efficient evaluation. It also reveals that while QA metrics generally improve over standard metrics that measure factuality across domains, performance is highly dependent on the way in which questions are generated.
  arXiv  Detail & Related papers  (2020-10-24T08:30:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.