Ziyong Wang, Yupei Lin, Runxuan Huang, Liang Lin, Dongyu Zhang, Guangrun Wang · 2026 · arXiv
DOI: 10.48550/arXiv.2610.28457 · open-access PDF
일반적인 목적의 멀티모델은 직접적인 로봇 행동 정책으로 약속을 보여주었지만, 실제 세계 상호 작용의 근본적인 물리학을 파악하고 그 이해를 행동을 안내하기 위해 적용할 수 있는지 여부는 아직 탐구되지 않았습니다. 우리는 대응 풍부한 조작을 대표적인 테스트베드로 사용하여이 광범위한 능력의 구체적인 예로 터치 이해를 검토하여이 질문을 조사합니다.
Machine translation (Meta M2M100); the English original follows.
General-purpose multimodal models have shown promise as direct robot action policies, yet whether they grasp the physics underlying real-world interactions and can apply that understanding to guide action remains underexplored. We investigate this question using contact-rich manipulation as a representative testbed, examining tactile understanding as a concrete instance of this broader capability. Contact-rich manipulation requires interpreting tactile feedback and adjusting actions after obstruction. Existing learned tactile policies acquire these capabilities through specialized pretraining
Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese
GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.28457, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txtSources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange