What is reproducibility in artificial intelligence and machine learning research?
Open accessAbstract
In the rapidly evolving fields of artificial intelligence (AI) and machine learning (ML), the reproducibility crisis underscores the urgent need for clear validation methodologies to maintain scientific integrity and encourage advancement. The crisis is compounded by the prevalent confusion over validation terminology. In response to this challenge, we introduce a framework that clarifies the roles and definitions of key validation efforts: repeatability, dependent and independent reproducibility, and direct and conceptual replicability. This structured framework aims to provide AI/ML researchers with the necessary clarity on these essential concepts, facilitating the appropriate design, conduct, and interpretation of validation studies. By articulating the nuances and specific roles of each type of validation study, we aim to enhance the reliability and trustworthiness of research findings and support the community's efforts to address reproducibility challenges effectively.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
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?
Some degree of run-to-run variability is unavoidable in deep learning frameworks, particularly when dealing with Language Models.
“recent works and official documentation emphasize that some degree of run-to-run variability is unavoidable in deep learning frameworks (Desai et al., 2025; Gundersen et al., 2018), in particular when dealing with Language Models (Semmelrock et al., 2025).”
From CiteFusion: an ensemble framework for citation intent classification harnessing dual-model binary couples and SHAP analyses (2025) · cites this paper as Desai et al., 2025
These factors are inherent to modern computational pipelines in deep learning.
“These factors are inherent to modern computational pipelines in deep learning (Desai et al., 2025; Goodman et al., 2016; Semmelrock et al., 2025)”
From CiteFusion: an ensemble framework for citation intent classification harnessing dual-model binary couples and SHAP analyses (2025) · cites this paper as Desai et al., 2025
Cites
1 retracted
Just, M. A., Pan, L., Cherkassky, V. L., McMakin, D. L., Cha, C., Nock, M. K., & Brent, D. (2017) · Nature Human Behaviour · doi:10.1038/s41562-017-0234-y · Cited by 210 · see citing works
Retracted1,500 scientists lift the lid on reproducibility
Baker, M. (2016) · Nature · doi:10.1038/533452a · Cited by 4,209 · see citing works
Zech, J. R., Badgeley, M. A., Liu, M., Costa, A. B., Titano, J. J., & Oermann, E. K. (2018) · PLOS Medicine · doi:10.1371/journal.pmed.1002683 · Cited by 1,719 · see citing works
Leakage and the reproducibility crisis in machine-learning-based science
Kapoor, S., & Narayanan, A. (2023) · Patterns · doi:10.1016/j.patter.2023.100804 · Cited by 1,387 · see citing works
What does research reproducibility mean?
Goodman, S. N., Fanelli, D., & Ioannidis, J. P. A. (2016) · Science Translational Medicine · doi:10.1126/scitranslmed.aaf5027 · Cited by 1,293 · see citing works
Measuring Massive Multitask Language Understanding
Hendrycks, D., Burns, C., Basart, S., Zou, A., Mazeika, M., Song, D., & Steinhardt, J. (2020) · arXiv · doi:10.48550/arxiv.2009.03300 · Cited by 291 · see citing works
Holistic Evaluation of Language Models
Liang, P., Bommasani, R., Lee, T., Tsipras, D., Soylu, D., Yasunaga, M., Zhang, Y., Narayanan, D., Wu, Y., Kumar, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2211.09110 · Cited by 255 · see citing works
Reproducibility of Benchmarked Deep Reinforcement Learning Tasks for Continuous Control
Islam, R., Henderson, P., Gomrokchi, M., & Precup, D. (2017) · arXiv · doi:10.48550/arxiv.1708.04133 · Cited by 177 · see citing works
Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference
Chiang, W.-L., Zheng, L., Sheng, Y., Angelopoulos, A. N., Li, T., Li, D., Zhang, H., Zhu, B., Jordan, M., Gonzalez, J. E., et al. (2024) · arXiv · doi:10.48550/arxiv.2403.04132 · Cited by 118
Flagged first, then by citations
Cited by
Sorted by citations
Paolini, L., Vahdati, S., Di Iorio, A., Wardenga, R., Heibi, I., & Peroni, S. (2025) · Scientometrics · doi:10.1007/s11192-025-05418-8 · Cited by 8 · see citing works
- Authors
- Desai, A., Abdelhamid, M., & Padalkar, N. R.
- Affiliations
- Ready Tensor, Inc. San Diego USA
- Boston University (US)
- Corresponding author
- Desai, Abhyuday
- Venue
- AI Magazine
- Publisher
- Wiley
- Published
- 2025-04-18
- Volume
- 46
- Issue
- 2
- ISSN
- 0738-4602 (ISSN-L)
- 2371-9621
- OpenAlex
- W4409574025
- Cited by
- 42
- References
- 9
- Citations by year
- 2026: 30 · 2025: 12
- FWCI
- 27.71
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- AI Magazine (publishedVersion, cc-by-nc)
- arXiv (Cornell University) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Topics
- Social Sciences › Decision Sciences › Information Systems and Management › Scientific Computing and Data Management
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
- Social Sciences › Safety Research › Ethics and Social Impacts of AI
- Keywords
- reproducibility, replication crisis, validation methodology, repeatability, research 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
