LATIDIA · Robótica

ACE-Brain-0.6: Scaling Spatial Grounding for Physical Foundation Models

arXiv:2610.27040v1 Tipo de anuncio: nuevo Resumen: Los modelos multimodales de propósito general, incluidos GPT-6 Astra y Claude Opus 5.5, demuestran amplias capacidades en la manipulación incorporada. Nuestro análisis indica que este gen

La noticia

arXiv:2610.27040v1 Announce Type: new Abstract: General-purpose multimodal models, including GPT-6 Astra and Claude Opus 5.5, demonstrate broad capabilities in embodied manipulation. Our analysis indicates that this generality is accompanied by strong spatial perception, as Astra can determine where to interact with task-relevant entities amid visual distractors. Because spatial perception connects high-level reasoning to the physical scene, we investigate scaling visual grounding to strengthen the perceptual foundation of general embodied operation. We develop a unified data pipeline that combines robot manipulation recordings with complementary visual data from diverse sources. After automatically annotating the robot recordings, the pipeline converts the collected data into verifiable grounding supervision. Fine-tuning Qwen3.5-9B on approximately 1.5 million