JEPA-Anything: One Predictive Framework Spans Vision, Biology, Weather and More
Taoyong Cui · hf · 2026-09-18
Gen-Verse introduces JEPA-Anything, a domain-agnostic world-modeling framework built on orthogonal predictive factorization (OPF), extending JEPA by decomposing latent targets into complementary factors learned via dedicated pathways.
- Evaluated across seven domains: vision, biology, clinical trajectories, control, molecular dynamics, physical fields, and weather
- Beats matched JEPA baselines on all 10 dynamics tasks; cuts single-intervention prediction error on Interventional Pong by 34.8%
- Lowest one-step and 100-step molecular errors across all four tested systems
- A factor-nominated biological intervention gained experimental support in cell co-cultures, organoids, tumor fragments, and mice; latent orbital modes recovered the Keplerian scaling exponent (slope -1.4991)
Results support a common factorized predictive principle linking world modeling with intervention and experimentally grounded scientific discovery. Code is open-sourced.
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