13 papers
Learning from Repaired Reasoning: Root-Cause-Guided On-Policy Distillation
Chenglei Shen, Haoyang Yao, Weijie Yu +3
On-policy self-distillation (OPSD) uses reference solutions as privileged hindsight to supervise student-generated reasoning trajectories. However, reference-based guidance may exp…
GenRecEdit: Adapting Model Editing for Generative Recommendation with Cold-Start Items
Chenglei Shen, Teng Shi, Weijie Yu +2
Generative recommendation (GR) has shown strong potential for sequential recommendation in an end-to-end generation paradigm. However, existing GR models suffer from severe cold-st…
Hybrid Orchestration of Edge AI and Microservices via Graph-based Self-Imitation Learning
Chen Yang, Jin Zheng, Yang Zhuolin +4
Modern edge AI applications increasingly rely on microservice architectures that integrate both AI services and conventional microservices into complex request chains with stringen…
Enhancing Bandit Algorithms with LLMs for Time-varying User Preferences in Streaming Recommendations
Chenglei Shen, Yi Zhan, Weijie Yu +2
In real-world streaming recommender systems, user preferences evolve dynamically over time. Existing bandit-based methods treat time merely as a timestamp, neglecting its explicit…
When Personalization Misleads: Understanding and Mitigating Hallucinations in Personalized LLMs
Zhongxiang Sun, Yi Zhan, Chenglei Shen +4
Personalized large language models (LLMs) adapt model behavior to individual users to enhance user satisfaction, yet personalization can inadvertently distort factual reasoning. We…
LLaDA-Rec: Discrete Diffusion for Parallel Semantic ID Generation in Generative Recommendation
Teng Shi, Chenglei Shen, Weijie Yu +6
Generative recommendation represents each item as a semantic ID, i.e., a sequence of discrete tokens, and generates the next item through autoregressive decoding. While effective,…
Balancing Stylization and Truth via Disentangled Representation Steering
Chenglei Shen, Zhongxiang Sun, Teng Shi +2
Generating stylized large language model (LLM) responses via representation editing is a promising way for fine-grained output control. However, there exists an inherent trade-off:…
Bridging Search and Recommendation through Latent Cross Reasoning
Teng Shi, Weicong Qin, Weijie Yu +4
Search and recommendation (S&R) are fundamental components of modern online platforms, yet effectively leveraging search behaviors to improve recommendation remains a challenging p…
Benefit from Rich: Tackling Search Interaction Sparsity in Search Enhanced Recommendation
Teng Shi, Weijie Yu, Xiao Zhang +3
In modern online platforms, search and recommendation (S&R) often coexist, offering opportunities for performance improvement through search-enhanced approaches. Existing studies s…
Similarity = Value? Consultation Value Assessment and Alignment for Personalized Search
Weicong Qin, Yi Xu, Weijie Yu +6
Personalized search systems in e-commerce platforms increasingly involve user interactions with AI assistants, where users consult about products, usage scenarios, and more. Levera…
MAPS: Motivation-Aware Personalized Search via LLM-Driven Consultation Alignment
Weicong Qin, Yi Xu, Weijie Yu +5
Personalized product search aims to retrieve and rank items that match users' preferences and search intent. Despite their effectiveness, existing approaches typically assume that…
Paragon: Parameter Generation for Controllable Multi-Task Recommendation
Chenglei Shen, Jiahao Zhao, Xiao Zhang +3
Commercial recommender systems face the challenge that task requirements from platforms or users often change dynamically (e.g., varying preferences for accuracy or diversity). Ide…
MoRE: A Mixture of Reflectors Framework for Large Language Model-Based Sequential Recommendation
Weicong Qin, Yi Xu, Weijie Yu +5
Large language models (LLMs) have emerged as a cutting-edge approach in sequential recommendation, leveraging historical interactions to model dynamic user preferences. Current met…