SAILRec: Steering LLM Attention to Dual-Side Semantically Aligned Collaborative Embeddings for Recommendation

• URL: http://arxiv.org/abs/2606.04514v1
• Date: Wed, 03 Jun 2026 06:46:32 GMT
• Title: SAILRec: Steering LLM Attention to Dual-Side Semantically Aligned Collaborative Embeddings for Recommendation
• Abstract summary: LLM-based recommenders enhance language models with collaborative embeddings from user-item interactions.<n>We find that the utilization of collaborative embeddings is depth-dependent and alignment-sensitive.<n>We propose SAILRec, an LLM-based recommender that improves this balance through dual-side semantic alignment and hierarchical attention steering.
• Score: 36.867366609555
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• Listen to the Latents: Self-Correcting Speech Recognition in Large Audio Language Models Through Hidden-State Interactions [120.64235456332916]
  Speech recognition systems increasingly integrate large language models (LLMs) to leverage their semantic knowledge.<n>We refine warm-adapted LLM-based ASR models by leveraging their own pre-adaptation base LLMs.<n>We propose Hybrid Search, a targeted correction strategy motivated by two observations.
  arXiv  Detail & Related papers  (2026-08-31T15:46:48Z)
• Harmonizing Semantic and Collaborative in LLMs: Reasoning-based Embedding Generator for Sequential Recommendation [6.3053631723979455]
  Large Language Models offer a promising way to enrich item semantics.<n>Current embedding methods fail to exploit the model's inner reasoning capacity.<n>We introduce ReaEmb, a novel framework that resolves both issues via a Latent Reasoning-enhanced Contrastive Learning stage and a Collaborative Reward Reinforcement Learning stage.
  arXiv  Detail & Related papers  (2026-06-15T13:37:52Z)
• AgentDR Dynamic Recommendation with Implicit Item-Item Relations via LLM-based Agents [42.177723613925146]
  We propose a novel LLM-agent framework, AgenDR, which bridges LLM reasoning with scalable recommendation tools.<n>Our approach delegates full-ranking tasks to traditional models while utilizing LLMs to integrate multiple recommendation outputs.<n>We show that our framework achieves superior full-ranking performance, yielding on average a twofold improvement over its underlying tools.
  arXiv  Detail & Related papers  (2025-10-07T05:48:05Z)
• LLM2Rec: Large Language Models Are Powerful Embedding Models for Sequential Recommendation [49.78419076215196]
  Sequential recommendation aims to predict users' future interactions by modeling collaborative filtering (CF) signals from historical behaviors of similar users or items.<n>Traditional sequential recommenders rely on ID-based embeddings, which capture CF signals through high-order co-occurrence patterns.<n>Recent advances in large language models (LLMs) have motivated text-based recommendation approaches that derive item representations from textual descriptions.<n>We argue that an ideal embedding model should seamlessly integrate CF signals with rich semantic representations to improve both in-domain and out-of-domain recommendation performance.
  arXiv  Detail & Related papers  (2025-06-16T13:27:06Z)
• Enhancing LLM-based Recommendation through Semantic-Aligned Collaborative Knowledge [25.757451106327167]
  SeLLa-Rec focuses on achieving alignment between the semantic spaces of Collabs. and LLMs.<n>This alignment fosters effective knowledge fusion, mitigating the influence of discriminative noise.<n> Experiments conducted on two public benchmark datasets demonstrate that SeLLa-Rec achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2025-04-14T11:15:30Z)
• Towards Scalable Semantic Representation for Recommendation [65.06144407288127]
  Mixture-of-Codes is proposed to construct semantic IDs based on large language models (LLMs) Our method achieves superior discriminability and dimension robustness scalability, leading to the best scale-up performance in recommendations.
  arXiv  Detail & Related papers  (2024-10-12T15:10:56Z)
• Enhancing High-order Interaction Awareness in LLM-based Recommender Model [3.7623606729515133]
  This paper presents an enhanced LLM-based recommender (ELMRec) We enhance whole-word embeddings to substantially enhance LLMs' interpretation of graph-constructed interactions for recommendations. Our ELMRec outperforms state-of-the-art (SOTA) methods in both direct and sequential recommendations.
  arXiv  Detail & Related papers  (2024-09-30T06:07:12Z)
• DaRec: A Disentangled Alignment Framework for Large Language Model and Recommender System [83.34921966305804]
  Large language models (LLMs) have demonstrated remarkable performance in recommender systems.<n>We propose a novel plug-and-play alignment framework for LLMs and collaborative models.<n>Our method is superior to existing state-of-the-art algorithms.
  arXiv  Detail & Related papers  (2024-08-15T15:56:23Z)
• Beyond Inter-Item Relations: Dynamic Adaption for Enhancing LLM-Based Sequential Recommendation [83.87767101732351]
  Sequential recommender systems (SRS) predict the next items that users may prefer based on user historical interaction sequences. Inspired by the rise of large language models (LLMs) in various AI applications, there is a surge of work on LLM-based SRS. We propose DARec, a sequential recommendation model built on top of coarse-grained adaption for capturing inter-item relations.
  arXiv  Detail & Related papers  (2024-08-14T10:03:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.