Six lessons from 60 papers: what a week of AI-for-Science research reveals
bravo_abad · x · 2026-10-19
The latest Discovery at Scale digest distills 60 AI-for-Science papers from the past week: an error-correction model learned on top of ECMWF improves 91/92/98% of pressure, temperature and precipitation targets; crystal text representations learn composition far more easily than atomic geometry; a policy ran 30 days and 3,000+ decisions on a robotic chemistry platform with a separate outcome predictor; 36% of 135 zinc protease designs worked experimentally; and bacteria maintain an analogue weight in plasmid abundance updated in-cell. Plus 38 shorter field-grouped notes.
More from Research
- Paras Chopra reviews Sensing Qualia: a fresh "external" take on the hard problem — paraschopra · 2026-10-19
- Anthropic gives patient $30,000 in Claude credits to research her own tumor — g3off_record · 2026-10-19
- PaperLeet open-sources a LeetCode for AI researchers: ~100 key papers, graded by tests — udmrzn · 2026-10-19
- Top probability journals shut submissions for a month citing surge and AI impact — michaelchchoi · 2026-10-19
- Bare IPython Harness Matches Production Agents on Qwen3.5-9B; Output Normalization Kills Run-to-Run Variance — HauntingMoment · 2026-10-19
- Bryan Kelly and Dacheng Xiu release 159-page Financial Machine Learning textbook free — mdancho84 · 2026-10-19