Disentangling the Properties of Human Evaluation Methods: A Classification System to Support Comparability, Meta-Evaluation and Reproducibility Testing
Open accessThe DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.
Abstract
Current standards for designing and reporting human evaluations in NLP mean it is generally unclear which evaluations are comparable and can be expected to yield similar results when applied to the same system outputs.This has serious implications for reproducibility testing and meta-evaluation, in particular given that human evaluation is considered the gold standard against which the trustworthiness of automatic metrics is gauged.Using examples from NLG, we propose a classification system for evaluations based on disentangling (i) what is being evaluated (which aspect of quality), and (ii) how it is evaluated in specific (a) evaluation modes and (b) experimental designs.We show that this approach provides a basis for determining comparability, hence for comparison of evaluations across papers, meta-evaluation experiments, reproducibility testing.
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
BLEURT: Learning Robust Metrics for Text Generation
Sellam, T., Das, D., & Parikh, A. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.704 · Cited by 1,005
BLEURT: Learning Robust Metrics for Text Generation
Sellam, T., Das, D., & Parikh, A. P. (2020) · arXiv · doi:10.48550/arxiv.2004.04696 · Cited by 1,005 · cited 2×
“…well with human judgements of specific quality criteria1 such as ‘Fluency,’ ‘Naturalness,’ ‘Readability’ or ‘Overall Quality’2 in the general case (Novikova et al., 2017; May and Priyadarshi, 2017; Reiter, 2018; Shimorina et al., 2018; Dušek et al., 2020; Sellam et al., 2020; Mathur et al., 2020).”
BackgroundFindings of the 2016 Conference on Machine Translation
Bojar, O., Chatterjee, R., Federmann, C., Graham, Y., Haddow, B., Huck, M., Jimeno Yepes, A., Koehn, P., Logacheva, V., Monz, C., et al. (2016) · Proceedings of the First Conference on Machine Translation: Volume 2, Shared Task Papers · doi:10.18653/v1/w16-2301 · Cited by 539
Why We Need New Evaluation Metrics for NLG
Novikova, J., Dušek, O., Cercas Curry, A., & Rieser, V. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1238 · Cited by 422 · cited 3×
“…well with human judgements of specific quality criteria1 such as ‘Fluency,’ ‘Naturalness,’ ‘Readability’ or ‘Overall Quality’2 in the general case (Novikova et al., 2017; May and Priyadarshi, 2017; Reiter, 2018; Shimorina et al., 2018; Dušek et al., 2020; Sellam et al., 2020; Mathur et…”
BackgroundWhy We Need New Evaluation Metrics for NLG
Novikova, J., Dušek, O., Curry, A. C., & Rieser, V. (2017) · arXiv · doi:10.48550/arxiv.1707.06875 · Cited by 406
A Structured Review of the Validity of BLEU
Reiter, E. (2018) · Computational Linguistics · doi:10.1162/coli_a_00322 · Cited by 386 · cited 3×
“More generalised conclusions are sometimes drawn, for example that BLEU scores do not correlate well with human judgements of specific quality criteria 1 such as ‘Fluency,’ ‘Naturalness,’ ‘Read-ability’ or ‘Overall Quality’ 2 in the general case (Novikova et al., 2017; May and Priyadarshi, 2017; Reiter, 2018; Shimorina…”
BackgroundBest 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×
“Human evaluations in NLG currently paint a confused picture3 with very poor standards for designing and reporting evaluations (van der Lee et al., 2019).”
BackgroundTowards Faithful Neural Table-to-Text Generation with Content-Matching Constraints
Wang, Z., Wang, X., An, B., Yu, D., & Chen, C. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.101 · Cited by 59
Automatic Poetry Generation from Prosaic Text
Van de Cruys, T. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.223 · Cited by 50 · cited 2×
“…(dotted underline), some just to one of the two, and others to neither (last three examples): “[whether] the poem [is] thematically structured” (Van de Cruys, 2020); “measures if a question is coherent with previous ones” (Chai and Wan, 2020); “measures ability of the dialogue sys-tem to produce…”
BackgroundYi, S., Goel, R., Khatri, C., Cervone, A., Chung, T., Hedayatnia, B., Venkatesh, A., Gabriel, R., & Hakkani-Tur, D. (2019) · Proceedings of the 12th International Conference on Natural Language Generation · doi:10.18653/v1/w19-8608 · Cited by 26
Cited by
All That’s ‘Human’ Is Not Gold: Evaluating Human Evaluation of Generated Text
Clark, E., August, T., Serrano, S., Haduong, N., Gururangan, S., & Smith, N. A. (2021) · Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) · doi:10.18653/v1/2021.acl-long.565 · Cited by 290
All That's 'Human' Is Not Gold: Evaluating Human Evaluation of Generated Text
Clark, E., August, T., Serrano, S., Haduong, N., Gururangan, S., & Smith, N. A. (2021) · arXiv · doi:10.48550/arxiv.2107.00061 · Cited by 290
The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics
Gehrmann, S., Adewumi, T., Aggarwal, K., Ammanamanchi, P. S., Aremu, A., Bosselut, A., Chandu, K. R., Clinciu, M.-A., Das, D., Dhole, K., et al. (2021) · Proceedings of the 1st Workshop on Natural Language Generation, Evaluation, and Metrics (GEM 2021) · doi:10.18653/v1/2021.gem-1.10 · Cited by 159
Self-citationXiao, C., Xu, S. X., Zhang, K., Wang, Y., & Xia, L. (2023) · Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023) · doi:10.18653/v1/2023.bea-1.52 · Cited by 138
Measuring Attribution in Natural Language Generation Models
Rashkin, H., Nikolaev, V., Lamm, M., Aroyo, L., Collins, M., Das, D., Petrov, S., Tomar, G. S., Turc, I., & Reitter, D. (2023) · Computational Linguistics · doi:10.1162/coli_a_00486 · Cited by 71
“Human evaluation studies are varied from paper to paper and detailed, reproducible annotation instructions are unavailable (Belz et al., 2020).”
Background
Metadata
- Authors
- Belz, A., Mille, S., & Howcroft, D. M.
- Venue
- Proceedings of the 13th International Conference on Natural Language Generation
- Published
- 2020
- Pages
- 183-194
- OpenAlex
- W3110131742
- Cited by
- 45
- References
- 10
- FWCI
- 4.03
- Type
- conference paper
- Language
- en
- Open access
- hybrid
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: Disentangling the Properties of Human Evaluation Methods: A Classification System to Support Comparability, Meta-Evaluation and Reproducibility Testing. https://app.cochat.ai/papers/10.18653/v1/2020.inlg-1.24
- The paper
- Belz, A., Mille, S., & Howcroft, D. M. (2020). Disentangling the Properties of Human Evaluation Methods: A Classification System to Support Comparability, Meta-Evaluation and Reproducibility Testing. Proceedings of the 13th International Conference on Natural Language Generation. https://doi.org/10.18653/v1/2020.inlg-1.24
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.