# Rethinking Latent Reasoning for Generative Recommendation: Beyond Teacher-Generated CoT and Recurrent Hidden-State Feedback

Authors: Yiwen Chen, Fuwei Zhang, Zehao Chen, Deqing Wang, Hehan Li, Peizhi Xu, Hanmeng Liu, Shuanglong Li, Xin Pei, Fuzhen Zhuang, Zhao Zhang.

Adds learned thought tokens between a user’s item history and the next-item prediction. Trains their hidden states to represent keywords from past items, then processes all thought tokens in one prompt pass without generating a written reasoning trace.

First public: 2026-11-06.
ArXiv initial submission: 2026-11-06.

[Original paper](https://arxiv.org/abs/2611.09006) · [PDF](https://arxiv.org/pdf/2611.09006) · [HTML](https://arxiv.org/html/2611.09006v1) · [Reading-list entry](https://latent-implicit-thinking.github.io/papers/2611.09006/)
[Preprint BibTeX](https://latent-implicit-thinking.github.io/papers/2611.09006/citation.bib) · [Preprint CSL-JSON](https://latent-implicit-thinking.github.io/papers/2611.09006/citation.csl.json) · [JSON record](https://latent-implicit-thinking.github.io/papers/2611.09006/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.

Reading-list version 0.15.2; literature checked through 2026-11-03.