Evaluation Gaps in Machine Learning Practice
Open accessThe DOI matches its Crossref record; 52 of the 52 references it cites were found in a registry or an index.
Abstract
Forming a reliable judgement of a machine learning (ML) model’s appropriateness for an application ecosystem is critical for its responsible use, and requires considering a broad range of factors including harms, benefits, and responsibilities. In practice, however, evaluations of ML models frequently focus on only a narrow range of decontextualized predictive behaviours. We examine the evaluation gaps between the idealized breadth of evaluation concerns and the observed narrow focus of actual evaluations. Through an empirical study of papers from recent high-profile conferences in the Computer Vision and Natural Language Processing communities, we demonstrate a general focus on a handful of evaluation methods. By considering the metrics and test data distributions used in these methods, we draw attention to which properties of models are centered in the field, revealing the properties that are frequently neglected or sidelined during evaluation. By studying these properties, we demonstrate the machine learning discipline’s implicit assumption of a range of commitments which have normative impacts; these include commitments to consequentialism, abstractability from context, the quantifiability of impacts, the limited role of model inputs in evaluation, and the equivalence of different failure modes. Shedding light on these assumptions enables us to question their appropriateness for ML system contexts, pointing the way towards more contextualized evaluation methodologies for robustly examining the trustworthiness of ML models.
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
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
Star, S. L., & Griesemer, J. R. (1989) · Social Studies of Science · doi:10.1177/030631289019003001 · Cited by 10,847
On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021) · Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency · doi:10.1145/3442188.3445922 · Cited by 6,988
Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author)
Breiman, L. (2001) · Statistical Science · doi:10.1214/ss/1009213726 · Cited by 4,340
GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
Wang, A., Singh, A., Michael, J., Hill, F., Levy, O., & Bowman, S. (2018) · Proceedings of the 2018 EMNLP Workshop BlackboxNLP: Analyzing and Interpreting Neural Networks for NLP · doi:10.18653/v1/w18-5446 · Cited by 4,222
Energy and Policy Considerations for Deep Learning in NLP
Strubell, E., Ganesh, A., & McCallum, A. (2019) · arXiv · doi:10.48550/arxiv.1906.02243 · Cited by 3,735
Energy and Policy Considerations for Deep Learning in NLP
Strubell, E., Ganesh, A., & McCallum, A. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1355 · Cited by 3,684 · cited 1×
“[40, 60, 159] and Was the labour paid fairly?”
Introduction to the CoNLL-2003 Shared Task: Language-Independent Named Entity Recognition
Tjong Kim Sang, E. F., & De Meulder, F. (2003) · Proceedings of the seventh conference on Natural language learning at HLT-NAACL 2003 - · doi:10.3115/1119176.1119195 · Cited by 2,788
Model Cards for Model Reporting
Mitchell, M., Wu, S., Zaldivar, A., Barnes, P., Vasserman, L., Hutchinson, B., Spitzer, E., Raji, I. D., & Gebru, T. (2019) · Proceedings of the Conference on Fairness, Accountability, and Transparency · doi:10.1145/3287560.3287596 · Cited by 2,681 · cited 4×
“In foregrounding the information needs of application developers, we are also aligned with calls for transparent reporting of ML model evaluations [117], prioritizing needs of ML fairness practitioners [77], model auditing practices [136], and robust practices for evaluating ML systems for…”
BackgroundInfluentialSelf-citationOn the Opportunities and Risks of Foundation Models
Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M. S., Bohg, J., Bosselut, A., Brunskill, E., et al. (2021) · arXiv · doi:10.48550/arxiv.2108.07258 · Cited by 2,676
Cited by
Mapping global dynamics of benchmark creation and saturation in artificial intelligence
Ott, S., Barbosa-Silva, A., Blagec, K., Brauner, J., & Samwald, M. (2022) · Nature Communications · doi:10.1038/s41467-022-34591-0 · Cited by 73
“Number of associated publications 14 (9-22) 2 (1-3) 0.”
BackgroundA Reasoning and Value Alignment Test to Assess Advanced GPT Reasoning
Mcintosh, T. R., Liu, T., Susnjak, T., Watters, P., & Halgamuge, M. N. (2024) · ACM Transactions on Interactive Intelligent Systems · doi:10.1145/3670691 · Cited by 42
Can We Trust AI Benchmarks? An Interdisciplinary Review of Current Issues in AI Evaluation
Eriksson, M., Purificato, E., Noroozian, A., Vinagre, J., Chaslot, G., Gomez, E., & Fernandez-Llorca, D. (2025) · arXiv · doi:10.48550/arxiv.2502.06559 · Cited by 35
de Koning, M., Machado, T., Ahonen, A., Strokina, N., Dianatfar, M., De Rosa, F., Minav, T., & Ghabcheloo, R. (2024) · Safety Science · doi:10.1016/j.ssci.2024.106517 · Cited by 17
“This involves evaluating our real-time safe pose estimator and assessing its performance, as discussed in Hutchinson et al. (2022).”
International AI Safety Report
Bengio, Y., Mindermann, S., Privitera, D., Besiroglu, T., Bommasani, R., Casper, S., Choi, Y., Fox, P., Garfinkel, B., Goldfarb, D., et al. (2025) · arXiv · doi:10.48550/arxiv.2501.17805 · Cited by 16
Towards Geographic Inclusion in the Evaluation of Text-to-Image Models
Hall, M., Bell, S. J., Ross, C., Williams, A., Drozdzal, M., & Soriano, A. R. (2024) · The 2024 ACM Conference on Fairness Accountability and Transparency · doi:10.1145/3630106.3658927 · Cited by 11
“Unlike the Region-CLIPScore, the Object-CLIPScore seems equally informative regardless of which location the annotator is in. Recommendation : Use a threshold for Object-CLIPScore to include multiple representations of the same concept.”
BackgroundCooper, N., Heldreth, C., & Hutchinson, B. (2024) · Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 2: Short Papers) · doi:10.18653/v1/2024.eacl-short.19 · Cited by 5
Self-citation
Metadata
- Authors
- Hutchinson, B., Rostamzadeh, N., Greer, C., Heller, K., & Prabhakaran, V.
- Venue
- 2022 ACM Conference on Fairness Accountability and Transparency
- Publisher
- ACM
- Published
- 2022-06-20
- Pages
- 1859-1876
- OpenAlex
- W4280621405
- Cited by
- 62
- References
- 52
- FWCI
- 10.79
- 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
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.