Hi~ I am Ke Fang, a PhD student in the Department of Computer Science and Engineering at the Pennsylvania State University, advised by Dr. Lu Cheng. Before that, I received my M.S. from the University of Pennsylvania and my B.E. from Tianjin University.

My research focuses on world models: learning predictive representations of how environments evolve, so that agents can understand, plan, and act. I am broadly interested in the challenges of building world models that are reliable and generalize across tasks. I have also worked on improving the reliability of large language models.

I am always happy to discuss research and potential collaborations. Feel free to reach out by email!

News

[2026.09] Our new paper, ATLAS: Aligned Transport of Latent Structure for Reliable World Model Planning, is now available on arXiv!

[2026.08] I started my PhD at the Pennsylvania State University.

Publications

• ATLAS: Aligned Transport of Latent Structure for Reliable World Model Planning
  Ke Fang, Yupu Yao, Lu Cheng. arXiv preprint, 2026. [arXiv]
• Rethinking Rubric-Based Reinforcement Learning through Process Supervision
  Ke Fang, Jiale Zhao, Lu Cheng. Under review, 2026.
• REHEARSE: Experiential Rehearsal for Verbal Confidence Calibration in Large Language Models
  Ke Fang, Tianyi Zhao, Qianwen Wang, Lu Cheng. arXiv preprint, 2025. [arXiv]