Metraipolis › archive › dc_ae419b8f62aa

Phase-and-First-Arrival VLM Feedback for Sparse-Reward Reinforcement Learning in Surgical Manipulation — arXiv:2609.07211

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.07211
License (as archived)
CC0-1.0
Archived
2026-10-16T15:31:04.750Z
SHA-256 of the archived text
1d49c64ae61fa98bf46a684052e79cfa543ca9ce0d529e310c4bb780cd664d53
Stable URL of this page
https://metraipolis.com/documento/dc_ae419b8f62aa

How to cite

Phase-and-First-Arrival VLM Feedback for Sparse-Reward Reinforcement Learning in Surgical Manipulation — arXiv:2609.07211 — arXiv (en), https://arxiv.org/abs/2609.07211 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ae419b8f62aa (Metraipolis archive dc_ae419b8f62aa, sha256 1d49c64ae61fa98bf46a684052e79cfa543ca9ce0d529e310c4bb780cd664d53)
BibTeX
@misc{dcae419b8f62aa,
  title = {Phase-and-First-Arrival VLM Feedback for Sparse-Reward Reinforcement Learning in Surgical Manipulation — arXiv:2609.07211},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.07211},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ae419b8f62aa; sha256 1d49c64ae61fa98bf46a684052e79cfa543ca9ce0d529e310c4bb780cd664d53},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07211. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Sparse outcome feedback limits what robots can learn from unsuccessful attempts at complex manipulation. Failed multi-stage surgical attempts can contain grasps, lifts, or transfers worth reusing. In sparse-reward reinforcement learning, terminal rewards collapse such attempts to the same outcome, while scalar vision-language model (VLM) ratings reveal neither what progress merits credit nor when it occurred. We introduce phase-and-first-arrival feedback: one VLM query per recorded episode identifies the furthest visually verified task phase and when that phase is first reached, allowing the learner to reuse partial behavior and localize credit. We instantiate it in SurgPhaseBench, a phase-structured suite spanning rigid and deformable tasks, and evaluate it in simulation and hardware. Across five simulated tasks, our method reaches 75.2% mean success, compared with 52.1% for a reward based on Contrastive Language-Image Pre-training (CLIP) using the same visual input; the advantage persists when only the feedback representation changes. On hardware, the same record supports autonomous block picking and slip recovery. Together, these results show that trajectory-level visual supervision can preserve partial progress while providing the temporal credit needed for sparse-reward control.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.RO).
Categorie arXiv: Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2609.07211 — 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 2026-10-16T15:11:47.575Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).