Knowledge and Data Engineering Group

Source code for ICML'26 paper "RQ-MoE: Residual Quantization via Mixture of Experts for Efficient Input-Dependent Vector Compression".

Source code for DASFAA'26 paper "GenPlugin: A Plug-and-Play Framework for Long-Tail Generative Recommendation with Exposure Bias Mitigation".

Source code for CIKM'25 paper: "SC-DAG: Semantic-Constrained Diffusion Attacks for Stealthy Exposure Manipulation in Visually-Aware Recommender Systems".

[ACL 2026 Findings] SWE-AGILE: A Software Agent Framework for Efficiently Managing Dynamic Reasoning Context

Source code for EMNLP'25 paper "CodeRAG: Finding Relevant and Necessary Knowledge for Retrieval-Augmented Repository-Level Code Completion".

[CVPR 2026 Findings] UI-AGILE: Advancing GUI Agents with Effective Reinforcement Learning and Precise Inference-Time Grounding

Source code for WSDM'26 paper: "HDRec: Hybrid Dual-Semantics Modeling for Enhancing LLM-Based Recommendation".

Source code for our TKDD paper "Deep Code Search with Naming-Agnostic Contrastive Multi-view Learning".

Source code for EMNLP 2024 Findings paper: Code Membership Inference for Detecting Unauthorized Data Use in Code Pre-trained Language Models.

Paper Reading list for Shilling Attack and Defense on Recommender Systems