Xiang Yue Publications



Xiang Yue (岳 翔)

Postdoctoral Researcher
Language Technologies Institute
School of Computer Science
Carnegie Mellon University

Email: xyue2@andrew.cmu.edu

Google Scholar Github Linkedin Twitter Wechat

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), where I was trained by Prof. Huan Sun and Prof. Yu Su. I also collaborate closely with Prof. Wenhu Chen at University of Waterloo.


My research aims to understand and enhance the reasoning capabilities of LLMs while improving their responsibility and reliability.
  • Understanding and enhancing the reasoning capabilities of (multimodal) LLMs
  • Code generation for autonomous programming and reasoning
  • Improving the responsibility of LLMs in areas such as privacy, robustness, and attribution.
  • LLMs for science and healthcare

Research Highlight

You might have heard of me because of the following work:
  • MMMU: the commonly used multimodal language model evaluation suite. 🔥[Slides]
  • MAmmoTH / MAmmoTH2: Strong reasoning models achieving SoTA in 2023 and 2024.
  • OpenCodeInterpreter: the open replication of OpenAI GPT-4 Code Interpreter to achieve SoTA on code generation tasks.
  • MixEval: a fast, cheap, and highly Arena correlated LLM benchmark.
  • MMLU-Pro: A more robust and challenging version of MMLU featuring more on reasoning.
  • Grokked Transformers: A mechanistic analysis revealing that why transformers could implicit reason through grokking.

What's New

Last Updated: 04/2024