Xiaoman Pan

PersonaAgent: When Large Language Model Agents Meet Personalization at Test Time

Soundness-Aware Level: A Microscopic Signature that Predicts LLM Reasoning Potential

Stabilizing Reinforcement Learning for Honesty Alignment in Language Models on Deductive Reasoning

DivScene: Benchmarking LVLMs for Object Navigation with Diverse Scenes and Objects

SePPO: Semi-Policy Preference Optimization for Diffusion Alignment