A Statistical Analysis of Summarization Evaluation Metrics Using Resampling Methods
Open accessThe DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.
Abstract
The quality of a summarization evaluation metric is quantified by calculating the correlation between its scores and human annotations across a large number of summaries. Currently, it is unclear how precise these correlation estimates are, nor whether differences between two metrics’ correlations reflect a true difference or if it is due to mere chance. In this work, we address these two problems by proposing methods for calculating confidence intervals and running hypothesis tests for correlations using two resampling methods, bootstrapping and permutation. After evaluating which of the proposed methods is most appropriate for summarization through two simulation experiments, we analyze the results of applying these methods to several different automatic evaluation metrics across three sets of human annotations. We find that the confidence intervals are rather wide, demonstrating high uncertainty in the reliability of automatic metrics. Further, although many metrics fail to show statistical improvements over ROUGE, two recent works, QAEval and BERTScore, do so in some evaluation settings.1
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
An analysis of variance test for normality (complete samples)
SHAPIRO, S. S., & WILK, M. B. (1965) · Biometrika · doi:10.1093/biomet/52.3-4.591 · Cited by 23,480
An Analysis of Variance Test for Normality (Complete Samples)
Shapiro, S. S., & Wilk, M. B. (1965) · Biometrika · doi:10.2307/2333709 · Cited by 23,469
BERTScore: Evaluating Text Generation with BERT
Zhang, T., Kishore, V., Wu, F., Weinberger, K. Q., & Artzi, Y. (2019) · arXiv · doi:10.48550/arxiv.1904.09675 · Cited by 2,865
Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond
Nallapati, R., Zhou, B., dos Santos, C., Gulcehre, C., & Xiang, B. (2016) · Proceedings of The 20th SIGNLL Conference on Computational Natural Language Learning · doi:10.18653/v1/k16-1028 · Cited by 2,361
Sample Size Requirements for Estimating Pearson, Kendall and Spearman Correlations
Bonett, D. G., & Wright, T. A. (2000) · Psychometrika · doi:10.1007/bf02294183 · Cited by 1,095
SummEval: Re-evaluating Summarization Evaluation
Fabbri, A. R., Kryściński, W., McCann, B., Xiong, C., Socher, R., & Radev, D. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00373 · Cited by 587 · cited 1×
“The standard method for calculating a CI for a correlation is the Fisher transformation (Fisher, 1992).”
MethodMoverScore: Text Generation Evaluating with Contextualized Embeddings and Earth Mover Distance
Zhao, W., Peyrard, M., Liu, F., Gao, Y., Meyer, C. M., & Eger, S. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1053 · Cited by 490
The Hitchhiker’s Guide to Testing Statistical Significance in Natural Language Processing
Dror, R., Baumer, G., Shlomov, S., & Reichart, R. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p18-1128 · Cited by 384
Self-citationRe-evaluating Evaluation in Text Summarization
Bhandari, M., Gour, P. N., Ashfaq, A., Liu, P., & Neubig, G. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.751 · Cited by 148
Cited by
Repairing the Cracked Foundation: A Survey of Obstacles in Evaluation Practices for Generated Text
Gehrmann, S., Clark, E., & Sellam, T. (2023) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.13715 · Cited by 123
Towards Question-Answering as an Automatic Metric for Evaluating the Content Quality of a Summary
Deutsch, D., Bedrax-Weiss, T., & Roth, D. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00397 · Cited by 75
Self-citationLongEval: Guidelines for Human Evaluation of Faithfulness in Long-form Summarization
Krishna, K., Bransom, E., Kuehl, B., Iyyer, M., Dasigi, P., Cohan, A., & Lo, K. (2023) · Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics · doi:10.18653/v1/2023.eacl-main.121 · Cited by 29
On the Limitations of Reference-Free Evaluations of Generated Text
Deutsch, D., Dror, R., & Roth, D. (2022) · arXiv · doi:10.48550/arxiv.2210.12563 · Cited by 29
Self-citationRe-Examining System-Level Correlations of Automatic Summarization Evaluation Metrics
Deutsch, D., Dror, R., & Roth, D. (2022) · arXiv · doi:10.48550/arxiv.2204.10216 · Cited by 22
Self-citation
Metadata
- Authors
- Deutsch, D., Dror, R., & Roth, D.
- Venue
- Transactions of the Association for Computational Linguistics
- Publisher
- MIT Press - Journals
- Published
- 2021
- Volume
- 9
- Pages
- 1132-1146
- ISSN
- 2307-387X (ISSN-L)
- OpenAlex
- W3211758383
- Cited by
- 46
- References
- 12
- FWCI
- 3.09
- Type
- journal article
- Language
- en
- Open access
- diamond
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: A Statistical Analysis of Summarization Evaluation Metrics Using Resampling Methods. https://app.cochat.ai/papers/10.1162/tacl_a_00417
- The paper
- Deutsch, D., Dror, R., & Roth, D. (2021). A Statistical Analysis of Summarization Evaluation Metrics Using Resampling Methods. Transactions of the Association for Computational Linguistics. https://doi.org/10.1162/tacl_a_00417
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.