Ph.D. student in Machine Learning (CSE)

Yizhou Zhao

Georgia Institute of Technology

I am a Ph.D. student in Machine Learning (CSE) at Georgia Tech, advised by Prof. Bo Dai.

My research spans post-training for reasoning and agents, self-evolving agentic systems, and agentic systems for robotics. I study how training signals and execution feedback can improve learning, adaptation, and transfer across tasks and environments. I have also worked on AI safety, including watermarking for language models and synthetic data.

I'm very fortunate to work closely with Prof. Weijie Su and Prof. Qi Long at the University of Pennsylvania, where I received my master's degree in Applied Mathematics and Computational Science (AMCS). I completed my bachelor's degree in Information and Computing Science at Zhejiang University.

Yizhou Zhao

News

Two papers have been accepted to NeurIPS 2026 (CurveRL and MarkTune)! See you in Atlanta!

I started my Ph.D. in Machine Learning (CSE) at Georgia Tech, advised by Prof. Bo Dai.

I completed my master's degree in AMCS at the University of Pennsylvania.

Our work on robust spectral watermarking for synthetic tabular data appeared in Statistical Learning and Data Science.

Research interests

I study learning and adaptation through model post-training, agent harnesses, and robot interaction.

Post-training for Reasoning and Agents

Reinforcement learning, on-policy distillation, and credit assignment for language model reasoning and agents. I study how training objectives, fine-grained signals, and context distributions shape learning efficiency and generalization in reasoning and interactive tasks.

Self-evolving Agentic Systems

Learnable harness evolution for recursive self-improvement, with updates to agent memory, tools, and orchestration learned from execution feedback. I study whether representing harness components as latent states helps these updates generalize across tasks and environments.

Agentic Systems for Robotics

Coding agents that write and execute programs to control robots through device interfaces for zero-shot task execution. I study how composable interfaces, reusable tools, and execution feedback support perception, planning, error recovery, and transfer across tasks and robot embodiments.

Selected publications

Full list on Google Scholar

* Equal contribution

2025

Robust Spectral Watermark for Synthetic Tabular Data

Yizhou Zhao, Xiang Li, Peter X. K. Song, Qi Long, Weijie Su

Statistical Learning and Data Science · Synthetic data · Watermarking

A post-editing watermark for synthetic tabular data that embeds signals in the frequency domain. The method handles heterogeneous features and uses rank-based pseudorandom bits for robustness to common data transformations.

2025

Toward Engineering AGI: Benchmarking the Engineering Design Capabilities of LLMs

Xingang Guo, Yaxin Li, Xiangyi Kong, …, Yizhou Zhao, …, Bin Hu

NeurIPS 2025 · Datasets & Benchmarks (Poster) · LLM evaluation · Engineering design

EngDesign evaluates language models on practical design tasks across nine engineering domains. Simulation-based evaluation tests whether generated designs meet functional goals and constraints, beyond factual recall or textbook questions.

Academic Service

Conference reviewer

  • NeurIPS 2026
  • ICLR 2027

Journal reviewer

  • Journal of the American Statistical Association (JASA)
  • The Annals of Applied Statistics (AOAS)
  • Harvard Data Science Review