LLMDiRec: LLM-Enhanced Intent Diffusion for Sequential Recommendation
Bo-Chian Chen, Manel Slokom · 2026
Read LLMDiRec: LLM-Enhanced Intent Diffusion for Sequential Recommendation by Bo-Chian Chen, Manel Slokom (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-2601-03259.txt.
Description
Bo-Chian Chen and Manel Slokom present LLMDiRec, a recommendation system that reads user intent from language, not just from item identifiers. Sequential recommenders predict what a user will want next from their history, but models built on ID embeddings lack semantic grounding and struggle with new users and rare, long-tail items. LLMDiRec mixes those collaborative ID signals with rich semantic representations produced by a large language model.
A dynamic fusion mechanism blends the two views, and a multi-task objective keeps them aligned while an intent-aware diffusion model generates the recommendation. Across five public datasets, the method outperformed leading algorithms. The gains were especially strong where intent is complex and for long-tail items, exactly the cases where ID-only representations carry the least information.
Catalog record
| CREATOR | Bo-Chian Chen, Manel Slokom |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-03259