{"ID":3052344,"CreatedAt":"2026-06-04T04:41:36.695875263Z","UpdatedAt":"2026-06-06T06:34:39.340662719Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.04514","arxiv_id":"2606.04514","title":"SAILRec: Steering LLM Attention to Dual-Side Semantically Aligned Collaborative Embeddings for Recommendation","abstract":"Recent LLM-based recommenders enhance language models with collaborative embeddings from user-item interactions, but making such embeddings available does not ensure their proper use during inference. Through a diagnostic attention analysis, we find that the utilization of collaborative embeddings is depth-dependent and alignment-sensitive, suggesting that LLMs need to balance their internal semantic knowledge with external collaborative knowledge. To address this issue, we propose SAILRec, an LLM-based recommender that improves this balance through dual-side semantic alignment and hierarchical attention steering. The former aligns item-side embeddings with item-text semantics and user-side embeddings with codebook-based semantic profiles, while the latter suppresses premature shallow-layer collaborative interference and strengthens collaborative evidence in deeper decision layers. Experiments on MovieLens-1M and Amazon-Book show that SAILRec consistently outperforms representative baselines, with ablation and masking analyses validating its key designs.","short_abstract":"Recent LLM-based recommenders enhance language models with collaborative embeddings from user-item interactions, but making such embeddings available does not ensure their proper use during inference. Through a diagnostic attention analysis, we find that the utilization of collaborative embeddings is depth-dependent an...","url_abs":"https://arxiv.org/abs/2606.04514","url_pdf":"https://arxiv.org/pdf/2606.04514v1","authors":"[\"Xi Wu\",\"Jiale Wang\",\"Zihan Wang\",\"Yichen Gao\",\"Xiaocui Yang\",\"Shi Feng\",\"Daling Wang\",\"Yifei Zhang\"]","published":"2026-06-03T06:46:32Z","proceeding":"cs.IR","tasks":"[\"cs.IR\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}