A medical-imaging audit found 13 of 59 papers using incorrect train-test splits
wandedob · x · 2026-07-27
The report classifies 59 in-scope papers and finds 26 correct splits, 13 incorrect splits, and 20 unclear cases.
It also says around 10 papers were excluded and roughly a dozen additional OCT/ultrasound leads could not be verified because of paywalls.
Related event: AI Agent Audit Reveals Data Leakage in Medical Imaging Papers(4 posts)→
More from Research
- Yale PhD student open-sources his paper figure scripts, packaged as a Skill for Claude Code and Cursor — burny_tech · 2026-09-23
- AI models now match superforecasters on ForecastBench; rematch set for October — burny_tech · 2026-09-23
- Dev uses Opus 5.5 with Lean to formally verify Claude Agent SDK, yielding 16 bug-fix PRs — bcherny · 2026-09-23
- Mathematicians, not just LLMs, made AI's math breakthroughs possible, scholars argue — tak3sh8 · 2026-09-23
- AI-enabled drug discovery cuts discovery time by 15-80%, McKinsey research finds — menhguin · 2026-09-23
- Gemini training details dissected: groupwise reward redistribution to fight reward hacking — nrehiew_ · 2026-09-23