19 papers
Reward Guided Decoding for Generative Recommendation
Ruochen Yang, Yusheng Huang, Youfeng Zheng +11
Generative recommendation formulates recommendation task into an SID sequence autoregressive generation paradigm, but the decoding process is often dominated by generation likeliho…
Unifying Generative Recall and Multi-Objective Ranking in a Single Decoder-Only Sequence
Ruochen Yang, Shuang Wen, Pengbo Xu +6
Modern industrial recommendation systems typically separate recall and ranking into two independent stages. Although this cascade supports corpus-level retrieval and fine-grained m…
Conditional Diffusion Guided Knowledge Transfer for Multi-Domain Knowledge Graph Completion
Jiawei Sheng, Taoyu Su, Xixun Lin +2
Multi-domain knowledge graph completion (MKGC) aims to improve missing triple prediction in a target KG by transferring knowledge from other support KGs. Existing methods typically…
Personalized Multi-Interest Modeling for Cross-Domain Recommendation to Cold-Start Users
Xiaodong Li, Jiawei Sheng, Jiangxia Cao +6
Cross-domain recommendation (CDR) has demonstrated to be an effective solution for alleviating the user cold-start issue. By leveraging rich user-item interactions available in a r…
Improving Reasoning Capabilities in Small Models through Mixture-of-Layers Distillation with Stepwise Attention on Key Information
Yao Chen, Jiawei Sheng, Wenyuan Zhang +1
The significant computational demands of large language models have increased interest in distilling reasoning abilities into smaller models via Chain-of-Thought (CoT) distillation…
HyperMem: Hypergraph Memory for Long-Term Conversations
Juwei Yue, Chuanrui Hu, Jiawei Sheng +5
Long-term memory is essential for conversational agents to maintain coherence, track persistent tasks, and provide personalized interactions across extended dialogues. However, exi…