New Cell Paper: World Models to Simulate Biomedical Interventions Before Running Them
marinkazitnik · x · 2026-09-18
ayushnoori announced the new paper "World models for biomedicine", published in Cell's special issue on AI in biology, co-led with njwfish.
The paper argues that simulating the consequences of an experiment before running it is the holy grail of AI for biology. In principle, biomedical world models could:
- simulate how biological systems respond to interventions never previously tested;
- search for interventions that drive cells, tissues, or patients toward desired states;
- plan experimental campaigns over long time horizons;
- power autonomous, closed-loop AI scientists.
Realizing this vision takes more than calling a model a "world model" — the paper examines what defines one and how it differs from related model classes.
Related event: Harvard team proposes biomedical world models framework in Cell(8 posts)→
More from Research
- ML is leaving its alchemy era: assumed truths can now be instrumented and falsified — mike64_t · 2026-09-18
- Together AI paper: score centering cancels drift in off-policy RL under train-inference mismatch — PandaAshwinee · 2026-09-18
- Cell's AI-in-biology special issue debuts LongevityBench, a 17-task aging benchmark — JosephJacks_ · 2026-09-18
- After OpenAI cracks Navier-Stokes, the era of the human mathematician is ending — yeastsplainer · 2026-09-18
- Custom BB-SLAM hits 0.02m accuracy with pure visual odometry, no LiDAR or IMU — broodsugar · 2026-09-18
- AI4Science Breakthrough Said to Exceed Expectations, Could Transform Computational Chemistry — BenBlaiszik · 2026-09-18