Stanford prof argues virtual cell challenge can't separate method gains from data advantage
anshulkundaje · x · 2026-11-20
Stanford professor Anshul Kundaje challenged claims coming out of the Arc Institute virtual cell competition: unless the dataset and evaluation are held constant across all methods, the challenge cannot validate a team's claimed methodological contribution, since data advantage and methodological superiority are fully confounded.
More from Research
- MIT's Hiroshi Ishii shares Tangible Media Group's Creative Tension principles — ishii_mit · 2026-11-20
- Train a real LLM entirely in your browser with this interactive demo — morgymcg · 2026-11-20
- Ex-OpenAI Researcher: English Is the New ML Programming Language, Lean the Type System — lukaszkaiser · 2026-11-20
- S1CogBench probes System One decision models across 24 cognitive tasks — Dengzhe Hou · 2026-11-20
- AFFECT from KAIST makes world models dynamics-faithful via action effect matching — John Won · 2026-11-20
- Mind2Act World: a 180-task benchmark exposing robots' mind-action composition failures — Mind2Act-World · 2026-11-20