ICML 2026 Oral Papers Show Poor Reproducibility; Researcher Vows Not to Hire Those with <50% Reproducible Papers
ChenhaoTan · x · 2026-08-06
A viral article on AI research reproducibility reveals that many ICML 2026 oral papers cannot be fully reproduced. In response, researcher Gabriel B. commits to never hiring anyone with over 50% of their papers not fully reproducible, based on GitHub issues and code release. This highlights growing concerns about reproducibility in AI research.
More from Research
- AI to Flood Math with New Results, Researchers Urge Profession to Adapt — TimothyDuignan · 2026-08-06
- COLM Paper Reveals Reasoning Faithfulness Limits in Vision-Language Models — nikaletras · 2026-08-06
- AI Struggles with Math Conjectures: Lacks Ability to Evaluate Problem Value — JFPuget · 2026-08-06
- Thomas Wolf Warns Against Separating Constitutional Training and RLVR Data Manifolds — Thom_Wolf · 2026-08-06
- Beihang University Unveils 2cm Micro-Bot Bug That Moves Fast and Senses Sound — lukas_m_ziegler · 2026-08-06
- Microsoft's PlugMem Slashes Agent Context by 100x via Knowledge Graphs — Roger_M_Taylor · 2026-08-06