A survey identified data leakage in 22 papers across 17 research fields, collectively affecting 294 papers.
Notes on verification
Directly confirmed by the original Kapoor & Narayanan (2023) Patterns paper text, its PMC full-text, and multiple independent citations (PubMed, ScienceDirect, arXiv) reproducing identical figures. [tier=gold indep_score=1.0 clusters=3 claim_tier=notable]
Sources
- Leakage and the reproducibility crisis in machine-learning-based science (seed:psychology_and_neuroscience)
- https://pmc.ncbi.nlm.nih.gov/articles/PMC10499856/ (corroboration)
- https://pubmed.ncbi.nlm.nih.gov/37720327/ (corroboration)
- https://www.sciencedirect.com/science/article/pii/S2666389923001599 (corroboration)
- https://arxiv.org/pdf/2602.17854 (corroboration)