LLM embeddings adapt progressively to improve item recommendations
PACERec: Progressive Adaptation of LLM Item Embeddings with Collaborative Evidence for Recommendation
Information RetrievalArtificial Intelligence
Summary
Recommendation systems sometimes struggle to suggest items that people like less often interact with, partly because they rely on limited data. The authors propose a new method called PACERec that adjusts large language model embeddings by learning how users typically move from one item to another. This approach helps the system better understand and recommend less popular items while keeping the original item meanings. Tests show it improves recommendations overall and especially for items with fewer interactions.
What this means in practice
- •For ecommerce platform developers: Enhance product recommendation systems to better suggest less popular items by improving item embedding adaptation with user interaction patterns.$Commercial implications: Enables the creation of commercial recommendation engines that increase sales and engagement by effectively promoting diverse products across catalogs.
- •For streaming service engineers: Improve sequential content recommendations by more accurately modeling transitions between watched items, especially for niche or rarely viewed content.
Authors
Junzhao Huang, Xi Wu, Zihan Wang, Zijing Wang, Ming Wang, Kang Liu, Xiaocui Yang, Shi Feng, Daling Wang, Yifei Zhang, Ting Yang, Xurui Li
Abstract
Large language model (LLM) embeddings have been increasingly adopted in sequential recommendation to encode static item attributes and alleviate the limitations of item ID embeddings under sparse interactions. However, these embeddings typically capture only static content similarity while neglecting sequential transitions, and since their adaptation quality relies on sparse per-item updates or global geometric calibration, existing methods still struggle to explicitly transfer preceding interaction contexts from frequent to long-tail items under unevenly distributed behavioral evidence, limiting recommendation effectiveness across the catalog. To address this challenge, we propose PACERec, an ID-free progressive adaptation framework that incorporates preceding interaction contexts into semantic representation adaptation by learning a shared velocity field from items with sufficient interactions via frequency-weighted flow matching. In this way, PACERec progressively guides static LLM embeddings toward collaborative behavioral targets while preserving original item semantics, overcoming sparse per-item updates to effectively generalize sequential transition patterns to long-tail items for both user modeling and candidate scoring. Extensive experiments on three real-world datasets across three sequential backbones demonstrate the superiority of our method over competitive baselines in both overall and long-tail recommendation.