LeCun co-organizes 'World Modeling for Physics' workshop at Aspen, Feb 2027
cosmo_shirley · x · 2026-10-02
A five-day 'World Modeling for Physics' workshop organized by Randall Balestriero, the Kempe Lab, Yann LeCun and others will run Feb 28–Mar 5, 2027 at the Aspen Center for Physics, with 70 participants and abstracts due Oct 5.
The premise: "the world is predictable in the right coordinates—the open question is whether a model can learn those coordinates on its own." Themes include JEPA and joint-embedding prediction, world models and latent-space dynamics/planning, and symmetry and physical priors (conservation laws, equivariance), asking whether self-supervised methods can recover physical structure from data and whether physical priors make world models more sample-efficient, interpretable, and reliable.
More from Research
- Adaptive effort is likely the killer app for dynamic looped transformers — willcb · 2026-10-02
- HuST Lab's Multimodal Flow: Fully Continuous Unified Language-Vision Generation — hustvl · 2026-10-02
- Kuaishou's DARA Cuts Multi-Reward RL Training Steps by Up to 65% — kuaishou · 2026-10-02
- Argo-Bench Pits Data Agents Against a 7.5B-Row Warehouse; Best Model Clears Only 34.8% of Tasks — textql · 2026-10-02
- LoopCD: training-free contrastive decoding lifts looped transformers, AIME 61.9%→73.3% — arankomatsuzaki · 2026-10-02
- Is Jev secretly learning a Value function? RL calibration and System 1/2 — lateinteraction · 2026-10-02