Wenjie Li

Supervised Fine-Tuning or Contrastive Learning? Towards Better Multimodal LLM Reranking

LIMI: Less is More for Agency

Refusal Falls off a Cliff: How Safety Alignment Fails in Reasoning?

Learning from the Best, Differently: A Diversity-Driven Rethinking on Data Selection

FoPO: Foresight Policy Optimization Incentivizes Strategic Reasoning in LLMs

Phased Training for LLM-powered Text Retrieval Models Beyond Data Scaling

Integrative Decoding: Improving Factuality via Implicit Self-consistency

R$^2$ec: Towards Large Recommender Models with Reasoning

LIMOPro: Reasoning Refinement for Efficient and Effective Test-time Scaling

SWIFT: On-the-Fly Self-Speculative Decoding for LLM Inference Acceleration

Direct Preference Optimization Using Sparse Feature-level Constraints

Constrain Alignment with Sparse Autoencoders

Subtle Errors Matter: Preference Learning via Error-injected Self-editing

TOMVALLEY: EVALUATING THE THEORY OF MIND REASONING OF LLMS IN REALISTIC SOCIAL CONTEXT