Truth or Error? Towards systematic analysis of factual errors in abstractive summaries
Open accessThe DOI matches its Crossref record; 6 of the 6 references it cites were found in a registry or an index.
Abstract
This paper presents a typology of errors produced by automatic summarization systems.The typology was created by manually analyzing the output of four recent neural summarization systems.Our work is motivated by the growing awareness of the need for better summary evaluation methods that go beyond conventional overlap-based metrics.Our typology is structured into two dimensions.First, the Mapping Dimension describes surface-level errors and provides insight into word-sequence transformation issues.Second, the Meaning Dimension describes issues related to interpretation and provides insight into breakdowns in truth, i.e., factual faithfulness to the original text.Comparative analysis revealed that two neural summarization systems leveraging pretrained models have an advantage in decreasing grammaticality errors, but not necessarily factual errors.We also discuss the importance of ensuring that summary length and abstractiveness do not interfere with evaluating summary quality.
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
Evaluating the Factual Consistency of Abstractive Text Summarization
Kryscinski, W., McCann, B., Xiong, C., & Socher, R. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2020.emnlp-main.750 · Cited by 669
Neural Text Summarization: A Critical Evaluation
Kryscinski, W., Keskar, N. S., McCann, B., Xiong, C., & Socher, R. (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/d19-1051 · Cited by 335
Durmus, E., He, H., & Diab, M. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.454 · Cited by 306
Falke, T., Ribeiro, L. F. R., Utama, P. A., Dagan, I., & Gurevych, I. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1213 · Cited by 265 · cited 4×
“Similarly, the authors of Falke et al. (2019) evaluate three different state-of-the-art systems and find that between 8 and 26% of the generated summaries contain at least one factual error, even though ROUGE scores indicate good performance.”
BackgroundResultInfluentialBest practices for the human evaluation of automatically generated text
van der Lee, C., Gatt, A., van Miltenburg, E., Wubben, S., & Krahmer, E. (2019) · Proceedings of the 12th International Conference on Natural Language Generation · doi:10.18653/v1/w19-8643 · Cited by 205 · cited 1×
“The ROUGE score is almost exclusively used as the optimization and evaluation metric in neural summarization methods, even though it has been recognized to be difficult to interpret and does not correlate well with human judgement (van der Lee et al., 2019).”
MethodAssessing The Factual Accuracy of Generated Text
Goodrich, B., Rao, V., Liu, P. J., & Saleh, M. (2019) · Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining · doi:10.1145/3292500.3330955 · Cited by 162 · cited 2×
“Additionally, Goodrich et al. (2019) compare several models such as relation extraction, binary classification and end-to-end models (E2E) for estimating factual accuracy on a Wikipedia text summarization task.”
BackgroundMethod
Cited by
QAFactEval: Improved QA-Based Factual Consistency Evaluation for Summarization
Fabbri, A., Wu, C.-S., Liu, W., & Xiong, C. (2022) · Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.18653/v1/2022.naacl-main.187 · Cited by 119
Annotating and Modeling Fine-grained Factuality in Summarization
Goyal, T., & Durrett, G. (2021) · Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.18653/v1/2021.naacl-main.114 · Cited by 100
Ladhak, F., Durmus, E., He, H., Cardie, C., & McKeown, K. (2022) · Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2022.acl-long.100 · Cited by 46
Santhanam, S., Hedayatnia, B., Gella, S., Padmakumar, A., Kim, S., Liu, Y., & Hakkani-Tur, D. (2021) · arXiv · doi:10.48550/arxiv.2110.05456 · Cited by 21
“Factual correctness has been studied in multiple generation tasks such as summarization (Maynez et al., 2020; Lux et al., 2020; Kryscinski et al., 2020; Nan et al., 2021; Zhou et al., 2020), document-level generation (Massarelli et al., 2020) and natural language generation (NLG) (Dušek and Kasner,…”
Background
Metadata
- Authors
- Lux, K.-M., Sappelli, M., & Larson, M.
- Venue
- Proceedings of the First Workshop on Evaluation and Comparison of NLP Systems
- Published
- 2020
- Pages
- 1-10
- OpenAlex
- W3103464296
- Cited by
- 14
- References
- 6
- FWCI
- 1.01
- Type
- conference paper
- Language
- en
- Open access
- gold
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: Truth or Error? Towards systematic analysis of factual errors in abstractive summaries. https://app.cochat.ai/papers/10.18653/v1/2020.eval4nlp-1.1
- The paper
- Lux, K.-M., Sappelli, M., & Larson, M. (2020). Truth or Error? Towards systematic analysis of factual errors in abstractive summaries. Proceedings of the First Workshop on Evaluation and Comparison of NLP Systems. https://doi.org/10.18653/v1/2020.eval4nlp-1.1
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.