Haitao Wang
2 papers · Latest:
Information Retrieval
DynamicPO: Dynamic Preference Optimization for Recommendation
DynamicPO prevents preference optimization collapse in LLM-based recommendation by adaptively selecting boundary negatives and adjusting optimization strength.
2605.00327
Information RetrievalSample Is Feature: Beyond Item-Level, Toward Sample-Level Tokens for Unified Large Recommender Models
SIF encodes full historical samples into tokens for unified large recommender models, improving performance and resolving feature heterogeneity.
2604.15650
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.