INTACT by ZJU & Tsinghua: Search-Free World Model Control Directly Outputs Actions
机器之心 · wechat · 2026-07-31
Zhejiang University, Tsinghua AIR and collaborating teams proposed INTACT (INtent-To-ACTion), a search-free world model control method based on end-to-end JEPA. It directly leverages states, actions, and future results from offline trajectories to translate motion intent into a semantic interface readable by the action model.
Traditional world models typically rely on CEM or MPPI at deployment to repeatedly predict and filter from a massive number of random candidate actions, incurring significant latency. INTACT introduces a shared conditional action operator, allowing the model to directly generate an action plan without searching candidate sequences, even in complex scenarios. Small-scale search is only introduced for verification in cases like fine contact.
Experiments show that on four official LeWM tasks, INTACT achieves an average success rate of 95.33% with zero search after training for only 1 epoch. With optional small-scale local verification, the success rate further improves to 96.86%. Compared to the traditional 9,000 candidate sequences, sampling volume is reduced by 23.44x, and planning latency is reduced to 1/300th.
Related event: INTACT: A Search-Free World Model by ZJU and Tsinghua(2 posts)→
More from Embodied
- New Review on Opportunities for Legged Robots by Jonas Frey et al. — ChongZzZhang · 2026-07-31
- Qlayers Robots Replace Hazardous Manual Labor, 6x Faster Tank Coating — lukas_m_ziegler · 2026-07-31
- 50+ Institutions Unite to Build Open-X-Tactile, World's Largest Tactile Dataset — RemiCadene · 2026-07-31
- Qualcomm Pushes for AI Gaming Era: On-device Compute Shrinks Dev Time to 5 Days — 智东西 · 2026-07-31
- Harness VLA: Boosting Frozen Robot Models Without Retraining — jiqizhixin · 2026-07-31
- Expert Reveals the Golden Rule of Robotics Demos: What You See Is What You Get — chris_j_paxton · 2026-07-31