Professor of Language Technologies Institute and Machine Learning Department
School of Computer Science of Carnegie Mellon University
Yiming
Yang
GHC 5703, LTI
Carnegie Mellon University
5000 Forbes Avenue
Pittsburgh, PA 15213-8213
Email: yiming AT cs
DOT cmu DOT edu
· LLM-Based Agents for Problem Solving: Research spans instruction tuning, demonstration-guided control, retrieval-augmented reasoning, and reinforcement learning—enabling adaptive agents for complex tasks such as math problem solving, combinatorial search, code synthesis and revision, and PDE solving.
· Combinatorial Optimization and AI for Science: Developed learning-based solvers for NP-hard problems using diffusion models, Langevin dynamics, and Fourier neural operators; contributed to neural PDE solving, scientific workload scheduling, and code-generation-driven scientific computing.
· Scalable Oversight and Learning Beyond Human Supervision: Advanced reinforcement learning–based alignment frameworks that scale supervision using self-play, instructible reward models, and principle-driven fine-tuning—enabling controllable and generalizable LLMs that go beyond human-written data and guidance.