# N0-VTLA: Scaling Vision-Tactile-Language-Action Model with Latent Tactile Tokens

arXiv: https://arxiv.org/abs/2607.23782 · HTML: https://mlsanity.com/paper/2607.23782/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Jul 2026.

## Built on by (most cited first, 6 of 6)

- [TacWAM: Anchor-Guided World Action Model with Mechanics-Aware Tactile Prediction](https://mlsanity.com/paper/2607.28391/) (2026)
- [VT-MUSE: Multimodal Unified Sequential Visuotactile Representation Learning for Manipulation](https://mlsanity.com/paper/2608.21290/) (2026)
- [HapticWAM: Distilling Imagined Touch into a World-Action Model without Inference-Time Tactile Sensing](https://mlsanity.com/paper/2609.23888/) (2026)
- [TacDyn-WAM: Learning Implicit Tactile Dynamics in a Heterogeneous Visuo-Tactile World Action Model](https://mlsanity.com/paper/2610.00638/) (2026)
- [TouchScale: 500 Hours of Human Vision and Touch for Visual-Tactile Learning](https://mlsanity.com/paper/2610.10288/) (2026)
- [Reinforcement Learning via Advantage Guidance](https://mlsanity.com/paper/2610.19783/) (2026)

## Abstract (arXiv)

We present N0-VTLA, a vision-tactile-language-action (VTLA) foundation model capable of (1) fine-grained contact-rich manipulation with tactile perception and tactile-feedback control, and (2) offline policy improvement from stored deployment data. Building on current vision-based backbones, we propose a training recipe for tactile integration consisting of visuo-tactile pre-training, staged tactile-pathway integration, and advantage-conditioned offline policy improvement. During pre-training, the policy learns broad contact priors from NeoData, our large-scale visuo-tactile robot dataset; to our knowledge, N0-VTLA is the first VTLA model pretrained on tactile data at scale. During post-training, we augment the policy with a predictive tactile pathway that distills the contact patterns learned at scale into the fine motion adjustments required by downstream tactile-centric manipulation. For offline policy improvement, we introduce ALTER, an advantage-conditioned offline reinforcement learning method that converts relative progress and trajectory-event comparisons into binary advantage labels for policy training on a fixed deployment corpus, further improving task-specific learning on contact-rich skills such as deformable object manipulation. Across contact-rich benchmarks, N0-VTLA outperforms strong baselines by wide margins: it wins all nine real-robot NeoReal tasks and reaches 63.8% mean success on a twenty-task simulation suite, against 44.0% for the strongest baseline. N0-VTLA policies trained with ALTER reach 75-95% success on three long-horizon real-robot tasks. These results lay a foundation for versatile tactile-driven manipulation policies.