Cell-sim contest winner admits its method is a denoised weighted mean, 2× over baseline
anshulkundaje · x · 2026-11-20
The author of the winning Arc Institute cell-simulation contest entry explains that at its core the method is a denoised weighted mean achieving roughly 2× the performance of a simple mean. Stanford professor Anshul Kundaje pushes back: such a baseline shortcut's effectiveness is entirely a function of train-test similarity and shouldn't count as a real "cell simulator."
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