{"slug": "omnibrain-0-a-unified-agentic-model-for-embodied-intelligence", "title": "OmniBrain-0: A Unified Agentic Model for Embodied Intelligence", "summary": "Researchers introduced OmniBrain-0, a lightweight unified embodied agent that coordinates manipulation and navigation through a shared high-level decision model, according to arXiv paper 2610.36013v1. OmniBrain-0 was trained with a progressive framework combining embodied pretraining, domain-specialized agent distillation and multi-teacher on-policy distillation (MOPD), plus Rubric-Guided Recursive Self-Improvement (RSI), a data flywheel in which a Meta-Agent converts execution evidence into refined rubrics, step-level reminders and verified trajectories. The model was evaluated on LIBERO-Pro, LIBERO-Plus, RoboCasa365, RoboTwin 2.0 and R2R-CE Val-Unseen and on multiple real-world manipulation tasks, combining strong cross-domain task performance with substantially lower inference and execution overhead than its closed-source teacher.", "body_md": "arXiv:2610.36013v1 Announce Type: new \nAbstract: General-purpose embodied intelligence requires robots to coordinate diverse skills over long horizons, track task progress, and recover from failures. Specialist policies alone do not ensure task-level reliability, while powerful multimodal agents rely on costly reasoning and external experience that compact models struggle to reproduce. We introduce OmniBrain-0, a lightweight unified embodied agent that coordinates manipulation and navigation through a shared high-level decision model, integrating task understanding, planning, progress tracking, and feedback-driven recovery. Our progressive training framework combines embodied pretraining, domain-specialized agent distillation, and multi-teacher on-policy distillation (MOPD). It distills powerful teachers' interaction experience into manipulation and navigation experts, then consolidates their decision capabilities into OmniBrain-0. Unification therefore occurs within the learned decision model rather than only at the tool interface. We further introduce Rubric-Guided Recursive Self-Improvement (RSI), a data flywheel in which a Meta-Agent turns execution evidence into refined rubrics, step-level reminders, and verified trajectories for domain-expert updates before their reconsolidation into OmniBrain-0 through MOPD. We evaluate OmniBrain-0 on LIBERO-Pro, LIBERO-Plus, RoboCasa365, RoboTwin 2.0, and R2R-CE Val-Unseen, as well as multiple real-world manipulation tasks. OmniBrain-0 combines strong cross-domain task performance with substantially lower inference and execution overhead than its closed-source teacher, demonstrating efficient transfer and consolidation of embodied capabilities into a compact unified agent.", "url": "https://wpnews.pro/news/omnibrain-0-a-unified-agentic-model-for-embodied-intelligence", "canonical_source": "https://arxiv.org/abs/2610.36013", "published_at": "2026-10-28 04:00:00+00:00", "updated_at": "2026-10-28 04:12:32.581377+00:00", "lang": "en", "topics": ["robotics", "artificial-intelligence", "ai-agents", "ai-research", "machine-learning"], "entities": ["OmniBrain-0", "LIBERO-Pro", "LIBERO-Plus", "RoboCasa365", "RoboTwin 2.0", "R2R-CE Val-Unseen", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/omnibrain-0-a-unified-agentic-model-for-embodied-intelligence", "markdown": "https://wpnews.pro/news/omnibrain-0-a-unified-agentic-model-for-embodied-intelligence.md", "text": "https://wpnews.pro/news/omnibrain-0-a-unified-agentic-model-for-embodied-intelligence.txt", "jsonld": "https://wpnews.pro/news/omnibrain-0-a-unified-agentic-model-for-embodied-intelligence.jsonld"}}