Metraipolis › archive › dc_775ea332a101

Imagine-RL: Residual-Confidence-Guided Cross-Attention for World-Model-Augmented VLA Reinforcement Learning — arXiv:2609.24033

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.24033
License (as archived)
CC0-1.0
Archived
2026-10-09T22:24:12.133Z
SHA-256 of the archived text
e3a2ed42b0c6836b721a22fb6c47b44cbc15c6ec40f435bb7f639c984505e10c
Stable URL of this page
https://metraipolis.com/documento/dc_775ea332a101

How to cite

Imagine-RL: Residual-Confidence-Guided Cross-Attention for World-Model-Augmented VLA Reinforcement Learning — arXiv:2609.24033 — arXiv (en), https://arxiv.org/abs/2609.24033 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_775ea332a101 (Metraipolis archive dc_775ea332a101, sha256 e3a2ed42b0c6836b721a22fb6c47b44cbc15c6ec40f435bb7f639c984505e10c)
BibTeX
@misc{dc775ea332a101,
  title = {Imagine-RL: Residual-Confidence-Guided Cross-Attention for World-Model-Augmented VLA Reinforcement Learning — arXiv:2609.24033},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.24033},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_775ea332a101; sha256 e3a2ed42b0c6836b721a22fb6c47b44cbc15c6ec40f435bb7f639c984505e10c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.24033. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reliable action evaluation in contact-rich manipulation requires looking beyond the current observation to future visual and contact consequences. Existing noise-space reinforcement learning efficiently steers a frozen Vision-Language-Action (VLA) policy, but its critics largely ignore these consequences. We present Imagine-RL, which augments noise-space VLA post-training with action-conditioned visual-torque imagination. For each candidate action chunk, a frozen visual-torque latent world model (VTLWM) autoregressively predicts compact future representations without pixel reconstruction. A current image-state-action query attends to observed histories and predicted futures, while previous-window prediction residuals provide token-wise confidence priors that suppress unreliable future tokens. By combining current evidence with predicted consequences, the action critic better evaluates candidate actions and supervises the actor, while the VLA and VTLWM remain frozen. Across four real-robot tasks with 50 evaluation trials per task, Imagine-RL uses only 100 RL trajectories and improves the average success rate by (23.6%) over DSRL and by (60%) over VLA baselines.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-22 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.24033 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-09T21:59:59.660Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).