Bridging Textual Profiles and Latent User Embeddings for Personalization
Zhaoxuan Tan, Xiang Zhai, Yan Zhu, Meng Jiang, Mohamed Hammad · 2026
Read Bridging Textual Profiles and Latent User Embeddings for Personalization by Zhaoxuan Tan, Xiang Zhai, Yan Zhu, Meng Jiang, Mohamed Hammad (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-06981.txt.
Description
Zhaoxuan Tan and four co-authors try to unite two ways of describing users in recommendation systems. Number-based user embeddings work well for retrieval but are hard to read. Written profiles are easy to read but hard to train for real usefulness in recommendations.
Their framework, BLUE, uses a language model to write profiles and an embedding model to score them as rewards. A second signal, based on predicting the next item, keeps profiles meaningful as well as useful. Tests on Amazon and Google review data show gains over strong baselines, clear cross-domain transfer, and better personal context for question answering.
Catalog record
| CREATOR | Zhaoxuan Tan, Xiang Zhai, Yan Zhu, Meng Jiang, Mohamed Hammad |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 97 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.06981