Xiang Yue

Bio

I am a postdoctoral researcher at CMU working with Prof. Graham Neubig on natural language processing (NLP) and large language models (LLMs). I obtained my Ph.D. from The Ohio State University (OSU-NLP-Group) in 2023, where I was trained by Prof. Huan Sun and Prof. Yu Su. I received my B.S. in Computer Science from Wuhan University in 2018.

My research aims to understand and enhance the reasoning capabilities of LLMs across different modalities and contexts while improving their responsibility and reliability.

• Understanding LLMs' reasoning through rigorous evaluation and benchmarking (e.g., MMMU/MMMU-Pro/MMLU-Pro, Demystifying Long CoT, GrokkedTransformer)
• Improving LLMs' reasoning capabilities with synthetic data generation (e.g., MAmmoTH/MAmmoTH2/MAmmoTH-VL, Pangea, OpenCodeInterpreter)
• Designing responsible and reliable AI models (e.g., Differential Privacy, Unlearning, Attribution, etc.) and their applications in healthcare (e.g., ECG Multimodal LLM (PULSE))

🎓 I am looking for (remote) interns/collaborators to work on exciting projects, including: • LLM / AI Agent: RL, Reasoning, Memory, Experience Learning
• LLM ML Foundations: Memorization/Forgetting, Generalization
• Evaluation and Benchmarking
• Applications: Medical MLLMs and Medical AI Agents
• Other topics you are passionate about! Please feel free to reach out if interested!

What's New

• [May 2025] 6 papers (3 main + 3 findings) have been accepted to ACL 2025!
• [May 2025] 3 papers have been accepted to ICML 2025!
• [April 2025] Demystifying long CoT reasoning in LLMs got a 🏆 Best Paper Award at ICLR 2025 Workshop on Foundation Models in the Wild!
• [Jan 2025] 6 papers have been accepted to ICLR 2025!
• [Dec 2024] ✈️ Attending NeurIPS 2024 and presenting four papers: MAmmoTH2, MixEval, MMLU-Pro, and Grokked Transformers! Ping me if you want a coffee chat!
• [Oct 2024] ✈️ Attending COLM 2024 and presenting VisualWebBench!
• [Sept 2024] ✈️ Attending Generative AI Rising Star Workshop at UMASS.
• [Jun 2024] ✈️ Attending CVPR 2024 and presenting MMMU [Slides]
• [May 2024] 6 papers (3 main + 3 findings) have been accepted to ACL 2024!
• [March 2024] Honored to receive the 2024 Carnegie Bosch Postdoctoral Fellowship and started my postdoc at CMU LTI!
• [Oct 2023] Two papers on "Evaluating Attribution" and "Evaluating Reasoning" accepted to EMNLP 2023
• [July 2023] Paper on Synthetic Text Generation with Differential Privacy received an Honorable Mention at ACL 2023
• [May 2023] Honored to receive the 2023 CSE Graduate Research Award and the 2023 College of Engineering Exemplary Graduate Student Researcher Award at Ohio State