# LaPR: Latent Preference Reasoning for Generative Personalized Retrieval

Authors: Hanze Guo, Shunyu Zhang, Jiakai Tang, Jingwei Zhuo, Yi Zhang, Xuanping Li, Xiao Zhou.

Compresses teacher-written retrieval steps into learned discrete slot codes. Predicts how many slots a query and user history need, then generates the slots and an item identifier without writing a natural-language rationale.

First public: 2026-09-15.

[Original paper](https://aclanthology.org/2026.emnlp-main.2255/) · [Reading-list entry](https://latent-implicit-thinking.github.io/papers/guo2026lapr/)
[BibTeX](https://latent-implicit-thinking.github.io/papers/guo2026lapr/citation.bib) · [CSL-JSON](https://latent-implicit-thinking.github.io/papers/guo2026lapr/citation.csl.json) · [JSON record](https://latent-implicit-thinking.github.io/papers/guo2026lapr/record.json)

Date note: Author-hosted preprint upload date. [Source](https://github.com/xiaozhoucam/xiaozhoucam.github.io/commit/b0cfb2efbc8615b5caa753c59368b2a2b94e237e).

## Published version

LaPR: Latent Preference Reasoning for Generative Personalized Retrieval. Hanze Guo, Shunyu Zhang, Jiakai Tang, Jingwei Zhuo, Yi Zhang, Xuanping Li, Xiao Zhou. Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, 2026.
[Publication](https://aclanthology.org/2026.emnlp-main.2255/) · [BibTeX](https://latent-implicit-thinking.github.io/papers/guo2026lapr/publication.bib) · [CSL-JSON](https://latent-implicit-thinking.github.io/papers/guo2026lapr/publication.csl.json)
[Publication record](https://aclanthology.org/2026.emnlp-main.2255/).

## Approaches

- [Internalized reasoning](https://latent-implicit-thinking.github.io/topics/internalized-reasoning/): Learn from worked solutions so that some or all of the reasoning can later happen without writing those steps. Training may use a separate teacher model or progressively remove written steps from the model being trained.
- [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.