Troubling Trends in Machine Learning Scholarship: Some ML papers suffer from flaws that could mislead the public and stymie future research.
Open accessThe DOI matches its Crossref record; 61 of the 61 references it cites were found in a registry or an index.
Abstract
Flawed scholarship threatens to mislead the public and stymie future research by compromising ML’s intellectual foundations. Indeed, many of these problems have recurred cyclically throughout the history of AI and, more broadly, in scientific research. In 1976, Drew McDermott chastised the AI community for abandoning self-discipline, warning prophetically that "if we can’t criticize ourselves, someone else will save us the trouble." The current strength of machine learning owes to a large body of rigorous research to date, both theoretical and empirical. By promoting clear scientific thinking and communication, our community can sustain the trust and investment it currently enjoys.
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.
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 says what the claim attributes to it.
In the citing paper
“Regarding the anthropomorphic language describing AI, its usage has been recognised as potentially detrimental in other AI sub-fields, including machine learning [87], brain-inspired AI [88] or art-generating AI [73].”
From Deanthropomorphising NLP: Can a language model be conscious? (2024) · cites this paper as [87]
The cited source says what the claim attributes to it.
In the citing paper
“Failing to identify the sources of empirical gains is a longstanding issue in ML and related fields [29, 16].”
From AI Agents That Matter (2024) · cites this paper as [29]
The cited source says what the claim attributes to it.
In the citing paper
“Failing to do so or misrepresenting these comparisons can lead to follow-up work that rests on false assumptions (Lipton & Steinhardt, 2019;Leech et al., 2024).”
From Open Problems in Mechanistic Interpretability (2025) · cites this paper as Lipton & Steinhardt, 2019
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
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 1×
“has limited relevance to the proposed learning algorithm, but might confer a sense of theoretical depth to readers. The ubiquity of this issue is evidenced by the paper introducing the Adam optimizer [35]. In the course of introducing an optimizer with strong empirical performance, it also offers a theorem regarding co…”
BackgroundFully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834
(2014) · Cited by 33,419
A decision-theoretic generalization of on-line learning and an application to boosting
Freund, Y., & Schapire, R. E. (1997) · Journal of Computer and System Sciences · doi:10.1006/jcss.1997.1504 · Cited by 20,466
Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.123 · Cited by 19,741 · cited 1×
“In contrast, claims of human-level performance in [29] are better-qualified to refer to the ImageNet classification task (rather than object recognition more broadly).”
BackgroundVisualizing and understanding convolutional networks
Zeiler, M. D., & Fergus, R. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10590-1_53 · Cited by 15,839 · cited 1×
“ive adversarial networks. This term first took root in deep learning in [79], which does address deconvolution, but was later over-generalized to refer to any neural architectures using upconvolutions [78, 50]. Such overloading of terminology can create lasting confusion. New machine learning papers referring to deconv…”
BackgroundDermatologist-level classification of skin cancer with deep neural networks
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017) · Nature · doi:10.1038/nature21056 · Cited by 14,362 · cited 2×
“Take, for example, the “dermatologist-level classification of skin cancer” reported in [21].”
BackgroundBatch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S. (2024) · arXiv (Cornell University) · doi:10.57702/o9raffed · Cited by 14,214
Estimating the reproducibility of psychological science
Open Science Collaboration (2015) · Science · doi:10.1126/science.aac4716 · Cited by 8,515
(2010) · Cited by 8,261
Research works printed but not linked: 15
- Bengio Y., Neural Networks: Tricks of the Trade, ed. G. Montavon, G. B. Orr, KR Müller, 437
- Choromanska A., 2015, Proceedings of the 18th International Conference on Artificial Intelligence and Statistics.
- Goodfellow I. J., 2015, Proceedings of the International Conference on Learning Representations.
- Henderson P., 2018, Proceedings of the 32nd Association for the Advancement of Artificial Intelligence Conference.
- Ioffe S., 2015, Proceedings of the 32nd International Conference on Machine Learning 37; http://proceedings.mlr.press/v37/ioffe15
- Kingma D. P., 2015, Proceedings of the 3rd International Conference on Learning Representations.
- Lipton Z. C., 2016, International Conference on Machine Learning Workshop on Human Interpretability.
- Lucic M., 2017, Proceedings of the 32nd Conference on Neural Information Processing Systems. arXiv Preprint arXiv:1711
- Melis G., 2018, Proceedings of the International Conference on Learning Representations.
- Reddi S. J., 2018, Proceedings of the International Conference on Learning Representations.
- Sculley D., 2018, Proceedings of the 6th International Conference on Learning Representations, Workshop Track.
- Steinhardt J., 2015, Proceedings of the 32nd International Conference on Machine Learning 37
- Steinhardt J., 2015, Proceedings of the 32nd International Conference on Machine Learning 37
- Szegedy C., 2013, International Conference on Learning Representations. arXiv Preprint arXiv:1312
- Zhang C., 2017, Proceedings of the International Conference on Learning Representations.
Cited by
On 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
Ray, P. P. (2023) · Internet of Things and Cyber-Physical Systems · doi:10.1016/j.iotcps.2023.04.003 · Cited by 2,463
Dau, H. A., Bagnall, A., Kamgar, K., Yeh, C.-C. M., Zhu, Y., Gharghabi, S., Ratanamahatana, C. A., & Keogh, E. (2019) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2019.1911747 · Cited by 1,128
“While this is a fictional report, note that all the numbers presented in the table and figure are true values, based on reproducible experiments that we performed [16].”
BackgroundMethodCan You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
Ovadia, Y., Fertig, E., Ren, J., Nado, Z., Sculley, D., Nowozin, S., Dillon, J. V., Lakshminarayanan, B., & Snoek, J. (2019) · arXiv · doi:10.48550/arxiv.1906.02530 · Cited by 989
“Contributions In the spirit of calls for more rigorous understanding of existing methods (Lipton and Steinhardt, 2018; Sculley et al., 2018; Rahimi and Recht, 2017), this paper provides a benchmark for evaluating uncertainty that focuses not only on calibration in the i.”
BackgroundMethodAn overview of deep learning in medical imaging focusing on MRI
Lundervold, A. S., & Lundervold, A. (2018) · NORA - Norwegian Open Research Archives · Cited by 1,012
Metadata
- Authors
- Lipton, Z. C., & Steinhardt, J.
- Affiliations
- Carnegie Mellon University (US)
- University of California, Berkeley (US)
- Venue
- Queue
- Published
- 2019-02
- Volume
- 17
- Issue
- 1
- Pages
- 45-77
- ISSN
- 1542-7730 (ISSN-L)
- 1542-7749
- OpenAlex
- W2881747041
- Cited by
- 231
- References
- 54
- Citations by year
- 2026: 30 · 2025: 14 · 2024: 17 · 2023: 20 · 2022: 21 · 2021: 35 · 2020: 46 · 2019: 34
- FWCI
- 159.63
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Queue (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Instituto de Ciencias del Mar y Limnología, Universidad Nacional Autónoma de México
- Topics
- Social Sciences › Library and Information Sciences › Academic Writing and Publishing
- Social Sciences › Library and Information Sciences › Academic Publishing and Open Access
- Social Sciences › Education › Academic integrity and plagiarism
- Keywords
- machine learning research, research integrity, scientific rigor, reproducibility, research critique, history of artificial intelligence, science communication
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: Troubling Trends in Machine Learning Scholarship: Some ML papers suffer from flaws that could mislead the public and stymie future research. https://app.cochat.ai/papers/10.1145/3317287.3328534
- The paper
- Lipton, Z. C., & Steinhardt, J. (2019). Troubling Trends in Machine Learning Scholarship: Some ML papers suffer from flaws that could mislead the public and stymie future research. Queue. https://doi.org/10.1145/3317287.3328534
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.