15 papers
Drafting Agents, Not Just Tokens: Speculative Recursive Self-Improvement for Agentic Systems
Xin Xu, Ke Cheng, Hao Wang +11
Improving an agent through successive revisions requires balancing exploration and evaluation under a limited budget. A narrow search can miss useful sequences of edits, while full…
What Should Data Teach? Moving Bottlenecks Across Circuit, Store, and Use
Yixiao Chen, Ke Cheng, Jiangtao Guan +5
What should data teach a language model at a particular point in training? A circuit view reveals three distinct bottlenecks: forming a computation, making its required content ava…
The User Asks, Platforms Compete: How Agentic Recommendation Markets Take Shape
Deyao Hong, Kehan Zheng, Qian Li +3
Online recommendation has traditionally taken place after a user enters a platform, which determines the candidate pool and the ranking shown to the user. LLM-based user agents ena…
Beyond Positive Signals: Unlocking Implicit Negative Behaviors for Enhanced Sequential User Modeling
Zexuan Cheng, Yue Liu, Jun Zhang +1
User behavior sequence modeling has become a central component in modern click-through rate (CTR) prediction. Over the past years, the community has invested substantial effort int…
S-GRec: Personalized Semantic-Aware Generative Recommendation with Asymmetric Advantage
Jie Jiang, Hongbo Tang, Wenjie Wu +6
Generative recommendation models sequence generation to produce items end-to-end, but training from behavioral logs often provides weak supervision on underlying user intent. Altho…
Reasoning to Rank: An End-to-End Solution for Exploiting Large Language Models for Recommendation
Kehan Zheng, Deyao Hong, Qian Li +4
Recommender systems are tasked to infer users' evolving preferences and rank items aligned with their intents, which calls for in-depth reasoning beyond pattern-based scoring. Rece…
Recurrent Preference Memory for Efficient Long-Sequence Generative Recommendation
Yixiao Chen, Yuan Wang, Yue Liu +9
Generative recommendation (GenRec) models typically model user behavior via full attention, but scaling to lifelong sequences is hindered by prohibitive computational costs and noi…
Spend Search Where It Pays: Value-Guided Structured Sampling and Optimization for Generative Recommendation
Jie Jiang, Yangru Huang, Zeyu Wang +4
Generative recommendation via autoregressive models has unified retrieval and ranking into a single conditional generation framework. However, fine-tuning these models with Reinfor…
Internalizing Multi-Agent Reasoning for Accurate and Efficient LLM-based Recommendation
Yang Wu, Haoze Wang, Qian Li +3
Large Language Models (LLMs) are reshaping recommender systems by leveraging extensive world knowledge and semantic reasoning to interpret user intent. However, effectively integra…
DiffuReason: Bridging Latent Reasoning and Generative Refinement for Sequential Recommendation
Jie Jiang, Yang Wu, Qian Li +6
Latent reasoning has emerged as a promising paradigm for sequential recommendation, enabling models to capture complex user intent through multi-step deliberation. Yet existing app…
ChainRec: An Agentic Recommender Learning to Route Tool Chains for Diverse and Evolving Interests
Fuchun Li, Qian Li, Xingyu Gao +7
Large language models (LLMs) are increasingly integrated into recommender systems, motivating recent interest in agentic and reasoning-based recommendation. However, most existing…
PRISM: Parallel Residual Iterative Sequence Model
Jie Jiang, Ke Cheng, Xin Xu +8
Generative sequence modeling faces a fundamental tension between the expressivity of Transformers and the efficiency of linear sequence models. Existing efficient architectures are…
End-to-End Semantic ID Generation for Generative Advertisement Recommendation
Jie Jiang, Xinxun Zhang, Enming Zhang +8
Generative Recommendation (GR) has excelled by framing recommendation as next-token prediction. This paradigm relies on Semantic IDs (SIDs) to tokenize large-scale items into discr…
Breaking the Curse of Repulsion: Remoteness-Aware Control of Negative Off-Policy Updates
Yusen Huo, Changping Wang, Yangru Huang +2
Off-policy policy optimization reuses historical behavior, including negative-advantage samples that suppress known failures. We show that repeated reuse can turn this useful signa…
SCoTER: Structured Chain-of-Thought Transfer for Enhanced Recommendation
Jie Jiang, Yang Wu, Qian Li +7
Harnessing the reasoning power of Large Language Models (LLMs) for recommender systems is hindered by two fundamental challenges. First, current approaches lack a mechanism for aut…