NeurIPS 2025: Published 'PolyMath: Evaluating Mathematical Reasoning in Multilingual Contexts' and 'Sampling-Efficient Test-Time Scaling'
ICLR 2025: Published 'Latent Space Chain-of-Embedding Enables Output-free LLM Self-Evaluation'
NeurIPS 2024: Published 'Embedding Trajectory for Out-of-Distribution Detection in Mathematical Reasoning'
ACL 2024: Published a paper on LLM reasoning
EMNLP 2024: Published two papers on summarization and LLM agents
ACL 2023: Published 'SumCoT' on LLM summarization
COLING 2022: Published a paper on low-resource summarization
Co-developed PolyMath, a multilingual mathematical reasoning benchmark adopted by Qwen3 for standard evaluation
Co-authored a survey in ACM Computing Surveys (IF=23.8): 'Igniting Language Intelligence: The Hitchhiker’s Guide From Chain-of-Thought Reasoning to Language Agents'