INTACT: Isomorphic Intent-to-Action Learning for Search-Free World Models
- Published
- Source
- arXiv
- Paper number
- 751
- Field
- Robotics
- arXiv ID
- 2607.26056
Key points
- It integrates a JEPA prediction architecture with an intent-to-action interface so it can generate actions directly without retrieval.
- It processes physical next states and target intents with one shared predictor, while designing the endpoint gradient asymmetrically.
- In direct mode without retrieval, it reaches 85.78% to 100% success on four LeWM tasks, with 2.9 to 5.5 ms inference time.
- With selective CEM search, it achieves 96.86% macro success while using 23.44x fewer samples than pure CEM.
- Joint training with a single encoder over four tasks yields 89.39% direct macro success.
Paper links
External research summaries. These are not HDATF publications or measured product results.