Google's Co-Scientist Expands to Multi-Disciplinary Closed-Loop Discovery
vivnat · x · 2026-09-01
Google DeepMind highlighted progress with Co-Scientist, extending it from biomedical hypothesis generation to execution-grounded closed-loop discovery across multiple scientific disciplines. Three new preprints showcase applications in math and cancer biology: In math, the AI collaborated iteratively with experts to tackle the Chowla sets problem; in cancer biology, Genentech introduced PerturbME, a sample-efficient method.
Related event: Google's Co-Scientist Achieves Closed-Loop Discovery Across Sciences(2 posts)→
More from Research
- AlphaEvolve sets new record for matrix multiplication exponent: omega to 2.371177 — 新智元 · 2026-09-01
- Building a long-term memory benchmark for agents: what to add? — True_Mongoose_7073 · 2026-09-01
- Building a high-recall, traceable "second brain" RAG system? — iMiguelmars · 2026-09-01
- Study reveals cross-layer activation patterns in hybrid attention models — 机器之心 · 2026-09-01
- Group-averaged Markov chains papers updated, blending group theory with Markov chains — michaelchchoi · 2026-09-01
- The agent doom loop isn't the model being dumb — it's the transcript working against you — RunAI_Coder · 2026-09-01