# Physical Words: A Vocabulary for Structuring Latent Reasoning in Robot Manipulation

Authors: Jiaming Liu, Xinghao Zhu, Yiming Zhang, Shanghang Zhang, Yinxi Wang, Yichu Yang, Zhuoyang Liu, Dantong Niu, Soha Pouya, Juekun Li, Yan Chang.

Pairs action-word predictions with hidden-state predictions of future visual and tactile observations. Uses these intermediate representations to guide robot manipulation, with separate supervision for action labels and future sensory information.

First public: 2026-10-19.
ArXiv initial submission: 2026-10-19.

[Original paper](https://arxiv.org/abs/2610.26854) · [PDF](https://arxiv.org/pdf/2610.26854) · [HTML](https://arxiv.org/html/2610.26854v3) · [Reading-list entry](https://latent-implicit-thinking.github.io/papers/2610.26854/)
[Preprint BibTeX](https://latent-implicit-thinking.github.io/papers/2610.26854/citation.bib) · [Preprint CSL-JSON](https://latent-implicit-thinking.github.io/papers/2610.26854/citation.csl.json) · [JSON record](https://latent-implicit-thinking.github.io/papers/2610.26854/record.json)

## Approaches

- [Thinking tokens](https://latent-implicit-thinking.github.io/topics/thinking-tokens/): Give a model extra tokens for computation, such as pauses, filler symbols, or learned planning tokens. These tokens need not spell out the solution.
- [Continuous-state reasoning](https://latent-implicit-thinking.github.io/topics/continuous-states/): Use vectors—lists of numbers—to carry intermediate reasoning from one step to the next, instead of selecting a text token at every step. Some methods use the model's hidden states; others combine several candidate token representations.

Reading-list version 0.11.0; literature checked through 2026-10-26.