Yasheng Wang

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

Advancing and Benchmarking Personalized Tool Invocation for LLMs

Distill Models by Aptitude: Efficient Reasoning Capability Distillation via Adaptive Data Curation and Overthinking Mitigation

InfoDeepSeek: Benchmarking Agentic Information Seeking for Retrieval-Augmented Generation

ATGen: Adversarial Reinforcement Learning for Test Case Generation

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

DeepDiver: Adaptive Web-Search Intensity Scaling via Reinforcement Learning

SPA-BENCH: A COMPREHENSIVE BENCHMARK FOR SMARTPHONE AGENT EVALUATION

QFFT, Question-Free Fine-Tuning for Adaptive Reasoning

Flat-LoRA: Low-Rank Adaptation over a Flat Loss Landscape

ToolACE: Winning the Points of LLM Function Calling

Bridging and Modeling Correlations in Pairwise Data for Direct Preference Optimization

RevisEval: Improving LLM-as-a-Judge via Response-Adapted References

Flat-LoRA: Low-Rank Adaption over a Flat Loss Landscape

Proactive Agent: Shifting LLM Agents from Reactive Responses to Active Assistance

Adaptive Tool Use in Large Language Models with Meta-Cognition Trigger

RethinkMCTS: Refining Erroneous Thoughts in Monte Carlo Tree Search for Code Generation

Learning Evolving Tools for Large Language Models