Metraipolis › archive › dc_a46d1d07ef74

Reinforcement Learning for Real-Time Vision-Language-Action Policies — arXiv:2609.18207

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.18207
License (as archived)
CC0-1.0
Archived
2026-10-15T15:07:56.622Z
SHA-256 of the archived text
821c16d0c82e4c541009354b1f362500c89b4a389c9a4da586d61a6f7935f041
Stable URL of this page
https://metraipolis.com/documento/dc_a46d1d07ef74

How to cite

Reinforcement Learning for Real-Time Vision-Language-Action Policies — arXiv:2609.18207 — arXiv (en), https://arxiv.org/abs/2609.18207 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a46d1d07ef74 (Metraipolis archive dc_a46d1d07ef74, sha256 821c16d0c82e4c541009354b1f362500c89b4a389c9a4da586d61a6f7935f041)
BibTeX
@misc{dca46d1d07ef74,
  title = {Reinforcement Learning for Real-Time Vision-Language-Action Policies — arXiv:2609.18207},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18207},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a46d1d07ef74; sha256 821c16d0c82e4c541009354b1f362500c89b4a389c9a4da586d61a6f7935f041},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18207. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement learning fine-tuning on top of large, pretrained Vision-Language-Action (VLA) models offers promise for highly reliable robot deployment. However, because of their scale, modern VLA models suffer from high inference latency, so the observation used to select an action is often stale by execution time, creating a distribution shift that can substantially degrade reliability and performance. Prior work has explored asynchronous policy execution to reduce the effect of latency, but these methods are mostly built on imitation learning and offer no mechanism for moving beyond the training distribution toward higher reliability. We close this gap by enabling RL fine-tuning that meets the real-time control requirements of dynamic real-world manipulation. Our approach builds on EXPO-FT, a framework for sample-efficient, reliable VLA fine-tuning with reinforcement learning, and decouples slow, expressive action generation from fast, reactive action edits: a large pretrained VLA proposes action chunks using its strong behavior prior, while a lightweight edit policy performs fast, reactive decision-making by editing actions in response to changes in state, conditioned on the latest observation. We instantiate this as Real-Time EXPO-FT, an RL framework for finetuning real-time VLA policies. On the Kinetix benchmark, Real-Time EXPO-FT enables a delayed policy to achieve the best performance among delayed and non-delayed methods in 10 out of 10 environments. On four dynamic real-world tasks, robot object passing, ball balancing, table soccer kicking, and dynamic object picking, with online robot data capped at 10 minutes, Real-Time EXPO-FT improves average policy performance from 42% to 97%, all without human intervention, demonstrating rapid, sample-efficient adaptation to challenging real-world dynamics. Website: this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.LG).
Categorie arXiv: Robotics (cs.RO) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.18207 — 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 2026-10-15T14:08:19.239Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).