Toward the Era of Agentic Embodied Intelligence
Toward the Era of Agentic Embodied Intelligence
This paper introduces A2-MEAS, an embodied multi-agent framework that uses code as a scaling interface to combine corrective language, spatial evidence, execution history, and action proposals.
The authors report benchmark results on public subsets of BEHAVIOR 2026 and RoboTwin 2.0, along with real-robot evaluations in household settings; claims about autonomy readiness and deployment status are attributed to the paper's assertions rather than independent verification.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
AVIOR 2026 public subset before online correction, 2.21x the 2025 winner's reported 0.2605 across different benchmark editions and training configurations. On all 100 BEHAVIOR 2026 public tasks, A2-MEAS achieves 0.6411 Q-score and 45.40% task success. On RoboTwin 2.0, A2-WLA achieves 93.98% average success, rising to 95.05% with the complete system. Across 12 real-robot tasks in five household settings, A2-WLA achiev
Open source S4
es 57.92% average success versus 26.67% for pi0.5. In long-horizon pour-over coffee preparation, a history-conditioned adapter raises post-failure completion at two filter-handling steps from 14% to 94% and from 8% to 82%, compared with A2-WLA retries. Our training infrastructure delivers 427.1 samples per second, 4.03x the throughput of a pi0.5 reference implementation on matched hardware. We will release the system
Open source S5
Source：arXiv Robotics — research abstracts · arxiv.org