让大模型更合理地使用用户-物品互动信息，提升推荐效果
SAILRec: Steering LLM Attention to Dual-Side Semantically Aligned Collaborative Embeddings for Recommendation
- 通过双侧语义对齐，让文本与交互嵌入更匹配
- 分层注意力控制，避免早期干扰、强化深层决策证据
- 在两个数据集上表现优于主流方法，验证设计有效性
基于大语言模型的推荐系统虽引入了用户-物品交互的协同嵌入，但其在推理中并未被有效利用。诊断性注意力分析发现，协同嵌入的使用具有深度依赖性和对齐敏感性，表明大模型需平衡内部语义知识与外部协同知识。为此，我们提出 SAILRec，通过双侧语义对齐和分层注意力引导，实现更优平衡：一方面将物品端嵌入与物品文本语义对齐，用户端嵌入与基于代码本的语义画像对齐；另一方面抑制浅层过早的协同干扰，增强深层决策层的协同证据。在 MovieLens-1M 与 Amazon-Book 上的实验表明，SAILRec 持续优于代表性基线，消融与掩码分析验证了其关键设计的有效性。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。