Reassessing automatic evaluation metrics for code summarization tasks
Open accessThe DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index.
Abstract
In recent years, research in the domain of source code summarization has adopted data-driven techniques pioneered in machine translation (MT). Automatic evaluation metrics such as BLEU, METEOR, and ROUGE, are fundamental to the evaluation of MT systems and have been adopted as proxies of human evaluation in the code summarization domain. However, the extent to which automatic metrics agree with the gold standard of human evaluation has not been evaluated on code summarization tasks. Despite this, marginal improvements in metric scores are often used to discriminate between the performance of competing summarization models. In this paper, we present a critical exploration of the applicability and interpretation of automatic metrics as evaluation techniques for code summarization tasks. We conduct an empirical study with 226 human annotators to assess the degree to which automatic metrics reflect human evaluation. Results indicate that metric improvements of less than 2 points do not guarantee systematic improvements in summarization quality, and are unreliable as proxies of human evaluation. When the difference between metric scores for two summarization approaches increases but remains within 5 points, some metrics such as METEOR and chrF become highly reliable proxies, whereas others, such as corpus BLEU, remain unreliable. Based on these findings, we make several recommendations for the use of automatic metrics to discriminate model performance in code summarization.
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
Figures
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.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The cited source doesn’t support the claim as written.
In the citing paper
“Roy et al. show that BLEU-4 improvements of more than +2 points are reasonable proxies for human-perceptible preference [18].”
From Few-shot training LLMs for project-specific code-summarization (2022) · cites this paper as [18]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe cited source doesn’t support the claim as written.
In the citing paper
“In most cases, we see improvements of over 2.0 BLEU, the required threshold for human perception noted by Roy et al. [57].”
From Automatic Semantic Augmentation of Language Model Prompts (for Code Summarization) (2024) · cites this paper as [57]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“BLEU [50] is the most widely-used, similarity-based measure for code summarization [57] and commit log generation [16].”
From Automatic Semantic Augmentation of Language Model Prompts (for Code Summarization) (2024) · cites this paper as [57]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Roy et al. [57] provide arguments, grounded on human subject study that for code summarization (our central task), that a gain of 2.0 or more BLEU is more likely to correspond with human perception of improvement.”
From Automatic Semantic Augmentation of Language Model Prompts (for Code Summarization) (2024) · cites this paper as [57]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“the 2-BLEU threshold noted by Roy et al. [57], below which BLEU results are unreliable predictors of human preference.”
From Automatic Semantic Augmentation of Language Model Prompts (for Code Summarization) (2024) · cites this paper as [57]
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
(1932) · Medical Entomology and Zoology · Cited by 9,568
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
A Call for Clarity in Reporting BLEU Scores
Post, M. (2018) · Proceedings of the Third Conference on Machine Translation: Research Papers · doi:10.18653/v1/w18-6319 · Cited by 2,112
A Call for Clarity in Reporting BLEU Scores
Post, M. (2018) · arXiv · doi:10.48550/arxiv.1804.08771 · Cited by 2,111
Proceedings of the 40th Annual Meeting on Association for Computational Linguistics
Isabelle, P. (2002) · Cited by 1,635
chrF: character n-gram F-score for automatic MT evaluation
Popović, M. (2015) · Proceedings of the Tenth Workshop on Statistical Machine Translation · doi:10.18653/v1/w15-3049 · Cited by 1,220
Murray, G., Renals, S. J., Carletta, J., & Moore, J. D. (2005) · Meeting of the Association for Computational Linguistics · Cited by 716
Summarizing Source Code using a Neural Attention Model
Iyer, S., Konstas, I., Cheung, A., & Zettlemoyer, L. (2016) · Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p16-1195 · Cited by 701
Hu, X., Li, G., Xia, X., Lo, D., & Jin, Z. (2018) · Proceedings of the 26th Conference on Program Comprehension · doi:10.1145/3196321.3196334 · Cited by 621
Using Machine Learning for Vulnerability Detection and Classification
Alon, U., Brody, S., Levy, O., & Yahav, E. (2018) · arXiv (Cornell University) · doi:10.4230/oasics.slate.2021.14 · Cited by 467
Research works printed but not linked: 13
- Miltiadis Allamanis , Hao Peng , and Charles Sutton . 2016 . A convolutional attention network for extreme summarization of source code . In International Conference on Machine Learning. 2091–2100 . Miltiadis Allamanis, Hao Peng, and Charles Sutton. 2016. A convolutional attention network for extreme summarization of source code. In International Conference on Machine Learning. 2091–2100.
- Uri Alon Shaked Brody Omer Levy and Eran Yahav. 2018. code2seq: Generating sequences from structured representations of code. arXiv preprint arXiv:1808.01400. Uri Alon Shaked Brody Omer Levy and Eran Yahav. 2018. code2seq: Generating sequences from structured representations of code. arXiv preprint arXiv:1808.01400. · arXiv
- Minghao Chen and Xiaojun Wan . 2019 . Neural Comment Generation for Source Code with Auxiliary Code Classification Task. In 2019 26th Asia-Pacific Software Engineering Conference (APSEC). 522–529 . Minghao Chen and Xiaojun Wan. 2019. Neural Comment Generation for Source Code with Auxiliary Code Classification Task. In 2019 26th Asia-Pacific Software Engineering Conference (APSEC). 522–529.
- Patrick Fernandes , Miltiadis Allamanis , and Marc Brockschmidt . 2018 . Structured Neural Summarization. In International Conference on Learning Representations. Patrick Fernandes, Miltiadis Allamanis, and Marc Brockschmidt. 2018. Structured Neural Summarization. In International Conference on Learning Representations.
- Tjalling Haije , Bachelor Opleiding Kunstmatige Intelligentie , E Gavves , and H Heuer . 2016 . Automatic comment generation using a neural translation model . Inf. Softw. Technol. , 55 , 3 (2016), 258 – 268 . Tjalling Haije, Bachelor Opleiding Kunstmatige Intelligentie, E Gavves, and H Heuer. 2016. Automatic comment generation using a neural translation model. Inf. Softw. Technol., 55, 3 (2016), 258–268.
- Yuding Liang and Kenny Zhu . 2018 . Automatic generation of text descriptive comments for code blocks . In Proceedings of the AAAI Conference on Artificial Intelligence. 32 . Yuding Liang and Kenny Zhu. 2018. Automatic generation of text descriptive comments for code blocks. In Proceedings of the AAAI Conference on Artificial Intelligence. 32.
- Chin-Yew Lin . 2004 . Rouge: A package for automatic evaluation of summaries. In Text summarization branches out. 74–81. Chin-Yew Lin. 2004. Rouge: A package for automatic evaluation of summaries. In Text summarization branches out. 74–81.
- L. Moreno , A. Marcus , L. Pollock , and K. Vijay-Shanker . 2013. JSummarizer: An automatic generator of natural language summaries for Java classes . In International Conference on Program Comprehension (ICPC). 230–232 . L. Moreno, A. Marcus, L. Pollock, and K. Vijay-Shanker. 2013. JSummarizer: An automatic generator of natural language summaries for Java classes. In International Conference on Program Comprehension (ICPC). 230–232.
- Stefan Riezler and John T Maxwell III. 2005 . On some pitfalls in automatic evaluation and significance testing for MT . In Proceedings of the ACL workshop on intrinsic and extrinsic evaluation measures for machine translation and/or summarization. 57–64 . Stefan Riezler and John T Maxwell III. 2005. On some pitfalls in automatic evaluation and significance testing for MT. In Proceedings of the ACL workshop on intrinsic and extrinsic evaluation measures for machine translation and/or summarization. 57–64.
- Peter Stanchev , Weiyue Wang , and Hermann Ney . 2020 . Towards a Better Evaluation of Metrics for Machine Translation . In Proceedings of the Fifth Conference on Machine Translation. Association for Computational Linguistics, Online. 928–933 . Peter Stanchev, Weiyue Wang, and Hermann Ney. 2020. Towards a Better Evaluation of Metrics for Machine Translation. In Proceedings of the Fifth Conference on Machine Translation. Association for Computational Linguistics, Online. 928–933.
- Akiyoshi Takahashi , Hiromitsu Shiina , and Nobuyuki Kobayashi . 2019 . Automatic Generation of Program Comments Based on Problem Statements for Computational Thinking. In 2019 8th International Congress on Advanced Applied Informatics (IIAI-AAI). 629–634 . Akiyoshi Takahashi, Hiromitsu Shiina, and Nobuyuki Kobayashi. 2019. Automatic Generation of Program Comments Based on Problem Statements for Computational Thinking. In 2019 8th International Congress on Advanced Applied Informatics (IIAI-AAI). 629–634.
- Bolin Wei Ge Li Xin Xia Zhiyi Fu and Zhi Jin. 2019. Code generation as a dual task of code summarization. In Advances in Neural Information Processing Systems. 6563–6573. Bolin Wei Ge Li Xin Xia Zhiyi Fu and Zhi Jin. 2019. Code generation as a dual task of code summarization. In Advances in Neural Information Processing Systems. 6563–6573.
- 10.5555/271581.271584 · doi:10.5555/271581.271584
Cited by
Few-shot training LLMs for project-specific code-summarization
Ahmed, T., & Devanbu, P. (2022) · Proceedings of the 37th IEEE/ACM International Conference on Automated Software Engineering · doi:10.1145/3551349.3559555 · Cited by 240
Geng, M., Wang, S., Dong, D., Wang, H., Li, G., Jin, Z., Mao, X., & Liao, X. (2024) · Proceedings of the IEEE/ACM 46th International Conference on Software Engineering · doi:10.1145/3597503.3608134 · Cited by 155
Source Code Summarization in the Era of Large Language Models
Sun, W., Miao, Y., Li, Y., Zhang, H., Fang, C., Liu, Y., Deng, G., Liu, Y., & Chen, Z. (2025) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1109/icse55347.2025.00034 · Cited by 213
Recommending Root-Cause and Mitigation Steps for Cloud Incidents using Large Language Models
Ahmed, T., Ghosh, S., Bansal, C., Zimmermann, T., Zhang, X., & Rajmohan, S. (2023) · 2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE) · doi:10.1109/icse48619.2023.00149 · Cited by 148
Out of the BLEU: how should we assess quality of the Code Generation models?
Evtikhiev, M., Bogomolov, E., Sokolov, Y., & Bryksin, T. (2023) · Journal of Systems and Software · doi:10.1016/j.jss.2023.111741 · Cited by 127
Metadata
- Authors
- Roy, D., Fakhoury, S., & Arnaoudova, V.
- Affiliations
- Washington State University (US)
- Venue
- Proceedings of the 29th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering
- Publisher
- ACM
- Published
- 2021-08-18
- Pages
- 1105-1116
- OpenAlex
- W3195727321
- Cited by
- 106
- References
- 52
- Citations by year
- 2026: 18 · 2025: 28 · 2024: 31 · 2023: 19 · 2022: 10
- FWCI
- 12.44
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE) (publishedVersion, cc-by)
- Topics
- Physical Sciences › Computer Science › Information Systems › Software Engineering Research
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- source code summarization, automatic evaluation metrics, human evaluation, BLEU, meteors, ROUGE, chrF, measurement reliability
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: Reassessing automatic evaluation metrics for code summarization tasks. https://app.cochat.ai/papers/10.1145/3468264.3468588
- The paper
- Roy, D., Fakhoury, S., & Arnaoudova, V. (2021). Reassessing automatic evaluation metrics for code summarization tasks. Proceedings of the 29th ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering. https://doi.org/10.1145/3468264.3468588
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.



