MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptation Article Swipe

MAKE IMPACT

Yi Xu , Moyu Zhang , Chenxuan Li , Zhihao Liao , Haibo Xing , Hao Deng , Jinxin Hu , Bowen Zhang , Xiaoyi Zeng , Jing Zhang · YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2508.15281

Recommender systems traditionally represent items using unique identifiers (ItemIDs), but this approach struggles with large, dynamic item corpora and sparse long-tail data, limiting scalability and generalization. Semantic IDs, derived from multimodal content such as text and images, offer a promising alternative by mapping items into a shared semantic space, enabling knowledge transfer and improving recommendations for new or rare items. However, existing methods face two key challenges: (1) balancing cross-modal synergy with modality-specific uniqueness, and (2) bridging the semantic-behavioral gap, where semantic representations may misalign with actual user preferences. To address these challenges, we propose Multimodal Mixture-of-Quantization (MMQ), a two-stage framework that trains a novel multimodal tokenizer. First, a shared-specific tokenizer leverages a multi-expert architecture with modality-specific and modality-shared experts, using orthogonal regularization to capture comprehensive multimodal information. Second, behavior-aware fine-tuning dynamically adapts semantic IDs to downstream recommendation objectives while preserving modality information through a multimodal reconstruction loss. Extensive offline experiments and online A/B tests demonstrate that MMQ effectively unifies multimodal synergy, specificity, and behavioral adaptation, providing a scalable and versatile solution for both generative retrieval and discriminative ranking tasks.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2508.15281

PDF

https://arxiv.org/pdf/2508.15281

OA Status

green

OpenAlex ID

https://openalex.org/W4416050722

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4416050722

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2508.15281

Digital Object Identifier

Title

MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptation

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2025

Year of publication

Publication date

2025-08-21

Full publication date if available

Authors

Yi Xu, Moyu Zhang, Chenxuan Li, Zhihao Liao, Haibo Xing, Hao Deng, Jinxin Hu, Bowen Zhang, Xiaoyi Zeng, Jing Zhang

List of authors in order

Landing page

https://arxiv.org/abs/2508.15281

Publisher landing page

PDF URL

https://arxiv.org/pdf/2508.15281

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2508.15281

Direct OA link when available

Cited by

0

Total citation count in OpenAlex

Full payload