N_0-TWAM: First Large-Scale Tactile-Native World-Action Model for Robots
NeoteAIEmbodied · hf · 2026-08-03
N0-TWAM is a tactile-native world-action model designed for contact-rich manipulation, capable of predicting both future vision and future contact states. It is the first known tactile world-action model trained at a large scale.
Key Highlights:
- Large-Scale Pre-training: Trained on visuo-tactile demonstrations spanning six embodiments and 450 tasks.
- Physical Grounding: Utilizes NeoForce, a unified force-based tactile representation, to condition action generation.
- Long-Horizon Optimization: Introduces tactile contact events for task staging to improve multi-stage manipulation.
- Efficient Architecture: Employs an asymmetric Mixture-of-Transformers architecture pairing a full-width expert for video prediction with slim experts for action and tactile prediction.
The codebase and model checkpoints will be open-sourced to foster further research in tactile-enabled robotic manipulation.
More from Embodied
- Honor Robot Phone Review: Mechanical Gimbal + AI Brain Turns Phones into Companions — 数字生命卡兹克 · 2026-08-03
- CS-JEPA: Decentralized Predictive Architecture for Swarm Robotics — ITMO · 2026-08-03
- ODEWorld: A Continuous-Time Latent World Model via ODE — Dongxiu Liu · 2026-08-03
- Comma.ai Recaps Hilarious Early Bugs in Self-Driving Models — dosco · 2026-08-03
- Interview with Luo Ping: The Core of Embodied AI Models is Bridging the Human-to-Robot Data Loop — 机器之心 · 2026-08-03
- HUST & Huawei's TurboVLA Bypasses LLM for 32Hz Real-Time Robot Control — 机器之心 · 2026-08-03