LATIDIA · Robótica

Open-HumanWAM: una receta abierta para escalar la experiencia humana y robótica para modelos de cimientos incorporados

arXiv: 2611.06817v2Tipo de anuncio: cross Resumen: El video humano egocéntrico ofrece una fuente escalable de antecedentes de manipulación para el aprendizaje de robots, pero traducir la escala de datos sin procesar en políticas de robots generalizables exige rigoro

La noticia

arXiv:2611.06817v2 Announce Type: cross Abstract: Egocentric human video offers a scalable source of manipulation priors for robot learning, yet translating raw data scale into generalizable robot policies demands rigorous data curation and training recipes. We introduce Open-HumanWAM, the first fully open recipe for scaling egocentric pretraining to 50,000 hours, anchored by Open-HumanWAM-50K---a curated egocentric video corpus densely annotated with hand motion and natural language. Leveraging this corpus, we systematically investigate pure human pretraining via closed-loop simulation evaluations and uncover an empirical log-linear scaling law linking pretraining data hours to downstream robot success rates. Grounded in these findings, we develop HumanWAM-Vanilla, a world--action model pretrained exclusively on 50k hours of egocentric