My name is George Ma (Jiangyan Ma). I am an EECS PhD student at UC Berkeley, advised by Prof. Somayeh Sojoudi. Previously, I was an undergraduate student at Peking University, where I did research on graph learning at Prof. Yisen Wang’s lab. My email: george_ma@berkeley.edu. My CV: George Ma’s CV. I actively write blogs on Zhihu: George M’s Zhihu Homepage.
Research Interests
- Interpretability of Large Language Models
- Code Generation
- Graph Neural Networks
- Invariant and Equivariant Neural Networks
Publications
Revising and Falsifying Sparse Autoencoder Feature Explanations
George Ma, Samuel Pfrommer, Somayeh Sojoudi (2025). Revising and Falsifying Sparse Autoencoder Feature Explanations. In Thirty-ninth Conference on Neural Information Processing Systems.
A Canonicalization Perspective on Invariant and Equivariant Learning
George Ma, Yifei Wang, Derek Lim, Stefanie Jegelka, Yisen Wang (2024). A Canonicalization Perspective on Invariant and Equivariant Learning. In Thirty-eighth Conference on Neural Information Processing Systems.
Baking Symmetry into GFlowNets
Jiangyan Ma, Emmanuel Bengio, Yoshua Bengio, Dinghuai Zhang (2023). Baking Symmetry into GFlowNets. In NeurIPS 2023 AI for Science: from Theory to Practice.
Laplacian Canonization: A Minimalist Approach to Sign and Basis Invariant Spectral Embedding
Jiangyan Ma, Yifei Wang, Yisen Wang (2023). Laplacian Canonization: A Minimalist Approach to Sign and Basis Invariant Spectral Embedding. In Thirty-seventh Conference on Neural Information Processing Systems.
Preprints
SpecAgent: A Speculative Retrieval and Forecasting Agent for Code Completion
George Ma, Anurag Koul, Qi Chen, Yawen Wu, Sachit Kuhar, Yu Yu, Aritra Sengupta, Varun Kumar, Murali Krishna Ramanathan (2025). SpecAgent: A Speculative Retrieval and Forecasting Agent for Code Completion. arXiv preprint arXiv:2510.17925.
Spooky Action at a Distance: Normalization Layers Enable Side-Channel Spatial Communication
Samuel Pfrommer, George Ma, Yixiao Huang, Somayeh Sojoudi (2025). Spooky Action at a Distance: Normalization Layers Enable Side-Channel Spatial Communication. arXiv preprint arXiv:2507.04709.
Professional Activities
- Top reviewer, NeurIPS 2024
- Reviewer, ICLR 2025
- Reviewer, AISTATS 2025
- Reviewer, ICML 2025
- Reviewer, NeurIPS 2025
- Reviewer, TMLR