Causal-JEPA: LeCun team proposes world models via object-level latent masking
burny_tech · x · 2026-08-29
The post discusses the Causal-JEPA paper (by Yann LeCun et al.), proposing an object-centric world model using latent masking. By masking object states and forcing inference from context during training, C-JEPA addresses interaction-dependent dynamics. It shows 20% improvement in counterfactual reasoning and uses only 1% of the input features for comparable performance in control tasks compared to patch-based models.
More from Research
- Pathway CSO pitches BDH as post-Transformer architecture uniting memory, reasoning, adaptation — schwentker · 2026-10-15
- EmbodiedEvo tops RoboDojo at 38.95% success by letting robots inherit failed practice — thetripathi58 · 2026-10-15
- 797GB of n-gram counts in CPU RAM let 1B models match 57% more compute — PangWeiKoh · 2026-10-15
- Google's HARNESS² evolves agent harnesses without verifiers, up to +11.5 points on WorkBuddy-Bench — dair_ai · 2026-10-15
- Subbarao Kambhampati to dissect the 'mythos of LLM thinking tokens' at UMN — rao2z · 2026-10-15
- How 3D Gaussian Splatting gives robots spatial memory of what's out of view — TheTuringPost · 2026-10-15