BBQRec: Behavior-Bind Quantization for Multi-Modal Sequential Recommendation

2025-04-09

AI Summary

Existing multimodal sequential recommendation methods suffer from two critical limitations when discretizing modalities into semantic IDs: (1) fragmented quantization decouples modality mapping from behavioral objectives, and (2) excessive reliance on discrete IDs undermines cross-modal semantic consistency, thereby weakening user preference modeling. To address these issues, we propose a behavior-anchored dual-alignment quantization framework—the first to jointly align behavioral and semantic spaces in contrastive codebook learning. We further design a discretization similarity reweighting mechanism grounded in quantized semantic relations, balancing modality synergy and model compatibility. Extensive experiments on four real-world benchmarks demonstrate significant improvements over state-of-the-art methods, validating that behavior-driven discretization enhances both recommendation accuracy and generalizability.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Multi-modal sequential recommendation systems leverage auxiliary signals (e.g., text, images) to alleviate data sparsity in user-item interactions. While recent methods exploit large language models to encode modalities into discrete semantic IDs for autoregressive prediction, we identify two critical limitations: (1) Existing approaches adopt fragmented quantization, where modalities are independently mapped to semantic spaces misaligned with behavioral objectives, and (2) Over-reliance on semantic IDs disrupts inter-modal semantic coherence, thereby weakening the expressive power of multi-modal representations for modeling diverse user preferences. To address these challenges, we propose a Behavior-Bind multi-modal Quantization for Sequential Recommendation (BBQRec for short) featuring dual-aligned quantization and semantics-aware sequence modeling. First, our behavior-semantic alignment module disentangles modality-agnostic behavioral patterns from noisy modality-specific features through contrastive codebook learning, ensuring semantic IDs are inherently tied to recommendation tasks. Second, we design a discretized similarity reweighting mechanism that dynamically adjusts self-attention scores using quantized semantic relationships, preserving multi-modal synergies while avoiding invasive modifications to the sequence modeling architecture. Extensive evaluations across four real-world benchmarks demonstrate BBQRec's superiority over the state-of-the-art baselines.

Problem

• Fragmented quantization misaligns modalities with behavioral objectives
• Over-reliance on semantic IDs disrupts inter-modal coherence
• Weak multi-modal representation limits diverse preference modeling

Innovation

• Behavior-semantic alignment for recommendation tasks
• Contrastive codebook learning to disentangle patterns
• Discretized similarity reweighting for self-attention adjustment