FORGE: Forming Semantic Identifiers for Generative Retrieval in Industrial Datasets

Kairui Fu*, Tao Zhang*, Shuwen Xiao*, Ziyang Wang, Xinming Zhang, Chenchi Zhang, Yuliang Yan, Junjun Zheng, Xiangheng Kong, Shengyu Zhang, Kun Kuang, Yuning Jiang

Hugging Face GitHub 知乎 WeChat

• Investigate what constitutes better semantic identifiers through a taxonomy of SID construction strategies and extensive downstream GR validation.
• Release AL-GR, an industrial-scale Taobao dataset with 14 billion interactions and multimodal features of 250 million items.
• Deployed on Taobao’s “Guess You Like” section, FORGE achieved a 0.35% increase in online transaction count.

RankGR: Rank-Enhanced Generative Retrieval with Listwise Direct Preference Optimization in Recommendation

Kairui Fu*, Changfa Wu*, Kun Yuan, Binbin Cao, Dunxian Huang, Yuliang Yan, Junjun Zheng, Jianning Zhang, Silu Zhou, Jian Wu, Kun Kuang

• Enhance generative retrieval with listwise DPO and lightweight rescoring to capture hierarchical preferences and richer user-item interactions.
• Deployed on Taobao’s “Guess You Like” section, RankGR achieved a 1.08% increase in online item page views and captured 49.88% of total exposures.

PI2I: A Personalized Item-Based Collaborative Filtering Retrieval Framework

Shaoqing Wang*, Yingcai Ma*, Kairui Fu*, Ziyang Wang, Dunxian Huang, Yuliang Yan, Jian Wu

Hugging Face

• A novel two-stage retrieval framework that enhances the personalization capabilities of traditional collaborative filtering.
• Deployed on Taobao’s “Guess You Like” section, PI2I achieved a 1.05% increase in online transaction rates.

ThinkRec: Thinking-based recommendation via LLM

Qihang Yu*, Kairui Fu*, Shengyu Zhang, Zheqi Lv, Fan Wu, Fei Wu

Project

• An early attempt to activate the reasoning ability of LLMs for more interpretable and effective recommendation.

RASTP: Representation-Aware Semantic Token Pruning for Generative Recommendation with Semantic Identifiers

Tianyu Zhan*, Kairui Fu*, Zheqi Lv, Shengyu Zhang

Project

• An effective strategy to selectively prune less informative tokens for semantic-identifier-based recommendation.

MALLOC: Benchmarking the Memory-aware Long Sequence Compression for Large Sequential Recommendation

Qihang Yu*, Kairui Fu*, Zhaocheng Du*, Yuxuan Si, Kaiyuan Li, Weihao Zhao, Zhicheng Zhang, Jieming Zhu, Quanyu Dai, Zhenhua Dong, Shengyu Zhang, Kun Kuang, Fei Wu

Project

• A benchmark that establishes a rigorous multi-dimensional evaluation protocol that couples standard ranking metrics with system-level constraints for long-sequence compression in large recommender systems.

CHORD: Customizing Hybrid-precision On-device Model for Sequential Recommendation with Device-cloud Collaboration

Tianqi Liu*, Kairui Fu*, Shengyu Zhang, Wenyan Fan, Zhaocheng Du, Jieming Zhu, Fan Wu, Fei Wu

• A device-cloud collaborative framework for personalized mixed-precision quantization that generates lightweight networks for heterogeneous mobile devices.

Tackling Device Data Distribution Real-time Shift via Prototype-based Parameter Editing

Zheqi Lv, Wenqiao Zhang, Kairui Fu, Qi Tian, Shengyu Zhang, Jiajie Su, Jingyuan Chen, Kun Kuang, Fei Wu

• Prototype-based parameter editing for tackling real-time distribution shifts on devices in both vision and recommendation tasks.

Forward Once for All: Structural Parameterized Adaptation for Efficient Cloud-coordinated On-device Recommendation

Kairui Fu, Zheqi Lv, Shengyu Zhang, Fan Wu, Kun Kuang

Project

• An early attempt at jointly customizing model structures and parameters for efficient cloud-coordinated on-device recommendation.

MergeNet: Knowledge Migration across Heterogeneous Models, Tasks, and Modalities

Kunxi Li*, Tianyu Zhan*, Kairui Fu*, Shengyu Zhang, Kun Kuang, Jiwei Li, Zhou Zhao, Fan Wu, Fei Wu

Project 知乎 WeChat

• Leverage parameters as the medium to achieve knowledge transfer across heterogeneous models, tasks, and modalities.

DIET: Customized Slimming for Incompatible Networks in Sequential Recommendation

Kairui Fu, Shengyu Zhang, Zheqi Lv, Jingyuan Chen, Jiwei Li

• Tackle parameter personalization and communication efficiency under strict device constraints in device-cloud collaborative recommendation.

End-to-End Optimization of Quantization-Based Structure Learning and Interventional Next-Item Recommendation

Kairui Fu, Qiaowei Miao, Shengyu Zhang, Kun Kuang, Fei Wu

• Investigate user distribution shifts in recommender systems and the difficulty of causal structure learning under recommender-system interventions.