Preserving Unstable Modes Through Inverse Dynamics in JEPA World Models

Leonardo F. Toso, Yann LeCun, James Anderson, Oumayma Bounou

arXivarxiv/2610.07540Oct 6, 2026cs.LGscore 0

Robotic systems often exhibit unstable modes, along which small perturbations and disturbances can cause unbounded growth unless corrected through feedback. Controlling such systems from high-dimensional visual observations requires representations that preserve these modes. Joint-embedding predictive architectures (JEPAs) provide a natural framework for learning such representations and their dynamics from visual data. However, we demonstrate that next step prediction combined with anti-collapse regularization does not guarantee that controllable unstable modes are preserved: the training…

SourceJSON

Comments and reviews

No comments yet.

API

curl /api/v1/records/arxiv/2610.07540
curl -X POST /api/v1/records/arxiv/2610.07540/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2610.07540/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'