LAW: Unifying Language, Action, and World Modeling for Agentic Physical Intelig
Overview
This research presents LAW, a closed-loop framework that combines vision-language models and world-action models to enable agentic physical intelligence. The system uses semantic subtask progress for planning, action evaluation, and execution monitoring. LAW achieves 94.21% average success on 50 RoboTwin tasks and 65% on real-robot tasks, outperforming existing methods like $pi_{0.5}$ and Fast-WAM.
From arXiv Robotics — research abstracts
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · Reported success rate: 95.56 percent · Basis not reported
“Dual-stream hierarchical visual memory maintains temporal context for both models, while subtask-local value learning aligns action evaluation with semantic objectives and preserves supervision from successful prefixes of failed trajectories. LAW achieves 94.21% average success across 50 RoboTwin tasks. Replacing the WAM backbone with OpenWAM-$alpha$ further raises the average success rate to 95.56%, establishing a n”
Exact source · revision 2Source owner not reported
Reported quantity · tasks: 50 other · Basis not reported · Differing source assertions
“Dual-stream hierarchical visual memory maintains temporal context for both models, while subtask-local value learning aligns action evaluation with semantic objectives and preserves supervision from successful prefixes of failed trajectories. LAW achieves 94.21% average success across 50 RoboTwin tasks. Replacing the WAM backbone with OpenWAM-$alpha$ further raises the average success rate to 95.56%, establishing a n”
Exact source · revision 2Source owner not reported
Reported quantity · tasks: 3 other · Basis not reported · Differing source assertions
“ew open-source state of the art on RoboTwin. Despite having only 2B parameters, our VLM also outperforms variants using Embodied-R1.5 and GPT-5.5 High. On three long-horizon real-robot tasks, LAW achieves 65% average success, compared with 28.33% for $pi_{0.5}$ and 25% for Fast-WAM.”
Exact source · revision 2Source owner not reported
Evidence dependency changes (3)
- 2026-10-30T22:34:05.712Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T22:33:50.017Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-27T20:37:45.795Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalLAW: Unifying Language, Action, and World Modeling for Agentic Physical Intelligence
This research presents LAW, a closed-loop framework that combines vision-language models and world-action models to enable agentic physical intelligence. The system uses semantic subtask progress for planning, action evaluation, and execution monitoring. LAW achieves 94.21% average success on 50 RoboTwin tasks and 65% on real-robot tasks, outperforming existing methods like $pi_{0.5}$ and Fast-WAM.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).