Hongshen Xu

DiSRouter: Distributed Self-Routing for LLM Selections

Empowering LLM Tool Invocation with Tool-call Reward Model

Task-Specific Data Selection for Instruction Tuning via Monosemantic Neuronal Activations

Enhancing LLM Reliability via Explicit Knowledge Boundary Modeling

Reducing Tool Hallucination via Reliability Alignment

Boosting Long-Context LLM Inference Efficiency with Intra-Layer Attention Similarity