Yuto Shibata, Kotaro Amaya, Kashu Yamazaki, Lalit Jayanti, Yoshimitsu Aoki, Mariko Isogawa, Katerina Fragkiadaki · 2026 · arXiv
DOI: 10.48550/arXiv.2610.28523 · open-access PDF
La interacción humano-humano requiere que los robots operen bajo severa observación parcial: durante el contacto físico cercano, el socio humano se oculta, truncado o fuera del campo de visión del robot, causando que la estimación de la posición falle y a menudo requiera un seguimiento externo o un estado humano privilegiado. Al mismo tiempo, las ocluciones mutuas pesadas introducen errores en los datos de captura de movimiento humano-humano utilizados para el aprendizaje. Presentamos REACT, un marco para aprender políticas de interacción humano-humano desplegables de demostraciones humano-humano.
Machine translation (Meta M2M100); the English original follows.
Humanoid-human interaction requires robots to operate under severe partial observability: during close physical contact, the human partner is occluded, truncated, or outside the robot's field of view, causing pose estimation to fail and often necessitating external tracking or privileged human state. At the same time, heavy mutual occlusions introduce errors into the human-human motion-capture data used for learning. We present REACT, a framework for learning deployable humanoid-human interaction policies from human-human demonstrations. REACT treats semantic contact structure as the key super
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=2610.28523, 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