I am a Computer Science & Math undergrad at MIT. In general, I’m passionate about making AI and computing systems safe and efficient so they can benefit humanity at large. Specifically, I am interested in topics of mechanistic interpretability, program synthesis, and performance engineering.

Experiences

I now try to understand the science of neural networks and interpret them in Prof. Max Tegmark’s lab. I have worked on measuring the algorithmic progress of language models and its role in AI scaling with Epoch AI and MIT FutureTech. Before that, I have researched neural compilation and SQL query compiler optimization in the Supertech Group.

Publications

Universal Neurons in GPT2 Language Models
_{Wes Gurnee, Theo Horsley, Zifan Carl Guo, Tara Rezaei Kheirkhah, Qinyi Sun, Will Hathaway, Neel Nanda, Dimitris Bertsimas. Preprint on arXiv.}

Measuring the Success of Diffusion Models at Imitating Human Artists
_{Stephen Casper*, Zifan Guo*, Shreya Mogulothu, Zachary Marinov, Chinmay Deshpande, Rui-Jie Yew, Zheng Dai, and Dylan Hadfield-Menell. Spotlighted in 2023 ICML Workshop on Generative AI and Law (GenLaw). (arXiv) (Twitter)}

Enabling Transformers to Understand Low-Level Programs
_{Zifan Carl Guo, and William Moses. In 2022 IEEE High Performance Extreme Computing Conference (HPEC).}

Contact

You can email me at carlguo[at]mit[dot]edu. You can also find me on GitHub, LinkedIn, or Twitter. You can find a time to chat here. Submit anonynous feedback to me here.

My pronouns are he/him/his. I can speak English and Chinese Mandarin.

Misc

Before coming to the US for high school, I spent most of my life in Jinan, China, where I’m happy to call home.

In Winter 2023, I competed in MIT Battlecode and won 2nd place (read more here). I used to compete in college-level Mock Trial, for which I coded a website for automatic tournament tabulation.

In my free time, I like to make coffee, listen to music on records, and run.


Live the questions now. Perhaps you will then gradually, without noticing it, live along some distant day into the answer. —Rainer Maria Rilke