Toward the Era of Agentic Embodied Intelligence
Abstract
Foundation models increasingly guide robot actions, yet sustained autonomy requires efficient control and failure recovery. We introduce A2-MEAS, an embodied multi-agent framework built around code as a scaling interface. Learned correction codes combine corrective language, spatial evidence, execution history, and action proposals. MetaBrain provides training supervision and runtime guidance. A correction encoder and adapter absorb this knowledge into policy weights during training, while fresh codes adapt behavior with frozen weights at deployment. The action model, A2-WLA, adds motion-aware history and a latent world model to a vision-language-action policy. Surprisingly, this simple design with action pretraining and task specialization reaches 0.5755 Q-score on the 50-task BEHAVIOR 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 achieves 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, models, training scripts, and infrastructure to support efficient embodied learning.
Figures & tables
| Agents | Responsibility | Function |
| Task verification | Maintain the task goal and verify execution progress | |
| Failure diagnosis | Review attempted actions and request missing evidence | |
| MetaBrain | Recovery instruction | Describe how the next attempt should change |
| Perceptual grounding | Identify objects, regions and visible properties | |
| Spatiotemporal grounding | Track identities, relative positions and changes over time | |
| Diagnostic agents | Interaction and feasibility | Assess contact, support and movement constraints |
| Setting | World model | VLM | Continuous action |
|---|---|---|---|
| Global batch | 4,608 | 2,048 | 2,048 |
| Per-GPU batch | 72 | 32 | 32 |
| Training plan | Up to 10 epochs | 150,000 steps | 500,000 steps |
| Save interval | 5,000 steps | 20,000 steps | 20,000 steps |
| Initial learning rate | |||
| Warm-up | 10,000 steps | 2,000 steps | 2,000 steps |
| Setting | Naive eight-GPU baseline | Optimized evaluation |
|---|---|---|
| Worker-pool GPUs | 8 (projected) | 8 Pro6000 GPUs |
| Workers per GPU | 1 | 3 |
| Concurrent workers | 8 | 24 |
| Evaluation episodes | 1,000 | 1,000 |
| Wall-clock duration | hours | hours |
| Episode throughput | per hour | per hour |
| Configuration | Q-score | Success (%) | Successes | Coverage |
|---|---|---|---|---|
| A 2 -WLA single model | 0.4597 | 27.20 | 272/1,000 | 70/100 |
| A 2 -WLA multiple models | 0.6040 | 40.60 | 406/1,000 | 80/100 |
| A 2 -MEAS complete system | 0.6411 | 45.40 | 454/1,000 | 88/100 |
| Team or submission | Institution | Q-score | SR (%) |
|---|---|---|---|
| George Albrecht | CAU / GMU / EWHA | 0.3862 | 20.00 |
| Mirua | Xiaomi Robotics | 0.3971 | 19.80 |
| VERA | AXXX | 0.4000 | 25.00 |
| Newbie | morphic & 4dvlab | 0.4040 | 21.70 |
| Huawei AI Algorithm Dept | Huawei Technologies | 0.4076 | 25.20 |
| Lion | Lionrock AI Lab | 0.4082 | 21.80 |
| Method | Q-score | SR (%) | Evaluation context |
| Competition public-dataset evaluation [ 17 ] | |||
| RLC [ 18 ] | 0.2605 | 11.20 | 4 checkpoints |
| Comet [ 17 ] | 0.1830 | 14.40 | Competition submission |
| SimpleAI Robot [ 17 ] | 0.1943 | 14.00 | Competition submission |
| The North Star [ 17 ] | 0.1702 | 12.80 | Competition submission |
| Post-competition and subsequent reports | |||
| Configuration | History frames | Latent world model | FAST init. | Macro SR (%) |
|---|---|---|---|---|
| (baseline) | — | — | — | 26.67 |
| A 2 -WLA configurations | ||||
| Config. 1 (base fine-tuning) | 2 | 39.58 | ||
| Config. 2 (+ latent world model) | 2 | 47.08 | ||
| Config. 3 (+ six-frame history) | 6 | 54.17 | ||
| A 2 -WLA (full configuration) | 6 | 57.92 | ||
| Setting | Model components | Training data | Performance | |||
| World model | Motion encoding | Skill sampling | Successful rollouts | Q-score | SR (%) | |
| Observation and initialization baselines | ||||||
| Current frame only | – | – | – | – | 0.1582 | 8.33 |
| Six-frame history | – | – | – | – | 0.2167 | 13.33 |
| FAST pretraining | – | – | – | – | 0.3453 | 20.83 |
| Action-policy initialization | – | – | – | – | 0.4846 | 39.72 |
| Model | Condition change | Mean Q | Success (%) |
|---|---|---|---|
| A 2 -WLA (w/o Adapter) | No Adapter | 0.5310 | 41.67 |
| Adapter w/o Language & Grounding | Remove correction language, task text and structured grounding | 0.5344 | 43.33 |
| Adapter w/o Long History | Retain only the latest state–action record | 0.5356 | 44.17 |
| Adapter w/o Candidate | Remove base-action candidate encoding | 0.5542 | 47.50 |
| Adapter w/o Grounding | Remove structured grounding input | 0.5570 | 49.17 |
| Full Adapter | All conditions retained | 0.6108 | 52.50 |
| Interface | Q-score | Physical recovery rate | Effective continuation rate | Final task completion rate |
|---|---|---|---|---|
| Instruction restatement (frozen A 2 -WLA) | 0.5668 | 46.27% | 25.37% | 38.81% |
| End-effector geometric correction (VLM + IK) | 0.6083 | 52.05% | 28.77% | 45.21% |
| Latent-space residual adapter | 0.6108 | 54.93% | 33.80% | 43.66% |
| Task | Task name | B: Q | F: Q | B: success | F: success |
|---|---|---|---|---|---|
| 000 | Turn on radio | 0.7000 | 0.7000 | 7/10 | 7/10 |
| 001 | Pick up trash | 0.6000 | 0.4667 | 5/10 | 3/10 |
| 006 | Hide Easter eggs | 0.3667 | 0.3778 | 3/10 | 3/10 |
| 017 | Bring water | 0.8333 | 0.3000 | 8/10 | 3/10 |
| 018 | Tidy bedroom | 0.7000 | 0.7333 | 5/10 | 3/10 |
| 022 | Put shoes on rack | 0.8500 | 0.7900 | 7/10 | 5/10 |
| Mode | Conditioning and scheduling | Q-score | SR (%) |
|---|---|---|---|
| F-VLM | Full task and online VLM-selected Focus | 0.4411 | 30.83 |
| F-Global | Full task only | 0.4201 | 30.00 |
| F-Linear | Simulator-assisted assessment, fixed order | 0.4115 | 25.00 |
| F-Adaptive | Same assessment, reordering and revisitation | 0.3785 | 27.50 |
VLMs for Robotics
Vision-language models for robotics connect visual and linguistic understanding to grounded navigation, manipulation, and robot control.
See all 700 papers