arXiv AI
Predictive Closure: A Set-Level Characterization of JEPA's Feature Selection
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Machine Learning
Jun 30
Predictive Objectives Discard Exogenous Control-Relevant Features: A Controlled Mechanistic Study
arXiv:2606. 30068v1 Announce Type: new Abstract: Joint-embedding predictive (JEPA-style) objectives learn representations by predicting future latents.
arXiv Machine Learning
Aug 27
JEPA-x: Cross-Predictive Physics Grounding for Forecastable Latent Dynamics
JEPA-x is a cross‑predictive physics grounding method that aligns visual latent dynamics with privileged physical trajectories. By treating visual observations and physical states as two views of the same action‑conditioned trajectory and sharing a predictor, it forces the model to learn a common transition rule for both modalities. The physical branch is only used during training, so deployment incurs no extra cost, and the approach significantly reduces rollout drift and boosts control success across a multi‑task suite.
arXiv AI
Aug 3
Auto-JEPA: A Latent World Model of Continuous Intent for End-to-End Autonomous Driving
arXiv:2607. 29031v1 Announce Type: cross Abstract: Existing autonomous-driving world models typically perform dense prediction of future videos, occupancy states, BEV representations, or agent motion.
arXiv AI
Oct 16
Learning Intrinsic Dynamics in Latent World Models
The paper investigates how encouraging action predictability in latent world models can preserve intrinsic dynamics, which are crucial for planning. By deriving a theoretical lower bound and conducting empirical tests, the authors show that action predictability prevents the encoder from merging distinct action outcomes, even when visual appearance varies. Using a method called Concat IDM, they achieve a substantial increase in planning success on the Push‑T benchmark, from 9.1% to 77.2%, demonstrating the practical benefit of this inductive bias.
arXiv Machine Learning
Oct 9
Refinement as a Service: Algorithmic Predictor Refinement
The paper investigates how to combine multiple calibrated predictors into a single refined predictor that preserves all available information and cannot be further refined. It models calibrated predictors as signaling schemes and defines refinement via feature‑independent garblings, showing that a new predictor is constructible exactly when its signal vector lies in the linear span of the input signal vectors. The authors provide algorithmic results: deterministic refinement is polynomial‑time solvable for two predictors but NP‑hard for many, while randomized refinement admits a polynomial‑time algorithm for any number of inputs by decomposing signal vectors into extreme rays of a polyhedral cone.
arXiv Machine Learning
Aug 4
Asleep at the Wheel: JEPA's Limitations in Evaluating Novel Driving Data
arXiv:2608. 01336v1 Announce Type: cross Abstract: Modern autonomous-driving fleets record far more video than human reviewers can inspect.