Metraipolis › archive › dc_bf8df7fdfcba

PF-RL: Progress Field Reinforcement Learning via Goal-Conditioned Value Geometry for Vision-Language-Action Models — arXiv:2609.32634

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.32634
License (as archived): CC0-1.0
Archived: 2026-10-08T00:16:57.117Z
SHA-256 of the archived text: e117f6dbf58891c811b277a9b1121defa2ce3fb8ee40e08b7149f3df5aa4275b
Stable URL of this page: https://metraipolis.com/documento/dc_bf8df7fdfcba

How to cite

PF-RL: Progress Field Reinforcement Learning via Goal-Conditioned Value Geometry for Vision-Language-Action Models — arXiv:2609.32634 — arXiv (en), https://arxiv.org/abs/2609.32634 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bf8df7fdfcba (Metraipolis archive dc_bf8df7fdfcba, sha256 e117f6dbf58891c811b277a9b1121defa2ce3fb8ee40e08b7149f3df5aa4275b)

BibTeX

@misc{dcbf8df7fdfcba,
  title = {PF-RL: Progress Field Reinforcement Learning via Goal-Conditioned Value Geometry for Vision-Language-Action Models — arXiv:2609.32634},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32634},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bf8df7fdfcba; sha256 e117f6dbf58891c811b277a9b1121defa2ce3fb8ee40e08b7149f3df5aa4275b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32634. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement Fine-Tuning~(RFT) has emerged as a promising paradigm for improving Vision-Language-Action~(VLA) policies, yet sparse task-level outcomes provide limited credit for intermediate transitions, especially in long-horizon manipulation. A natural approach is to model intermediate task progress and use it as dense feedback for policy improvement. Despite their architectural differences, existing progress-aware methods commonly formulate task progress as an explicit scalar prediction, providing limited structure for modeling how intermediate observations relate to the task goal, which may hinder effective transition-level credit assignment. We introduce Progress Field Reinforcement Learning (PF-RL), which learns a structured goal-conditioned progress representation over pretrained VLA features and converts it into dense credit for policy optimization. A lightweight shared Progress Field head maps current and goal representations into a compact progress space, where geometric distance induces goal-conditioned value, while complementary temporal and goal-structure objectives shape the learned geometry. Transition-level value changes naturally yield dense progress advantages, enabling fine-grained credit assignment for both offline policy improvement and online reinforcement fine-tuning. Extensive experiments on LIBERO, RoboTwin2.0, and real-world bimanual manipulation tasks show that PF-RL consistently improves policy performance over strong supervised fine-tuning, reinforcement fine-tuning, and progress-aware baselines.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.32634 — 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 2026-10-07T23:52:16.424Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).