Qun Liu

ToolACE-MT: Non-Autoregressive Generation for Agentic Multi-Turn Interaction

Teaching LLMs According to Their Aptitude: Adaptive Switching Between CoT and TIR for Mathematical Problem Solving

RidgeLoRA: Matrix Ridge Enhanced Low-Rank Adaptation of Large Language Models

ToolACE: Winning the Points of LLM Function Calling

ReAttention: Training-Free Infinite Context with Finite Attention Scope

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

EMOVA: Empowering Language Models to See, Hear and Speak with Vivid Emotions

Mixture of insighTful Experts (MoTE): The Synergy of Thought Chains and Expert Mixtures in Safety Self-Alignment

Roadmap towards Superhuman Speech Understanding using Large Language Models

LongSafetyBench: Long-Context LLMs Struggle with Safety Issues