{Zizhao, Zoe, Chen, 陈梓昭, 梓昭} are equally good with me.
I like {natural, programming, mathematical} languages, because they help us formalize this mushy and magical process in our brain that we call reasoning. These keywords catch my attention: grounding, generalization, compositionality, and creativity.
I am a CS PhD student at Cornell Tech, working with Yoav Artzi (2024-). Previously I studied robotics as an undergraduate in Toronto and worked there too.
Publications
- Chen, Z., Artzi, Y. (2025). Knot So Simple: A Minimalistic Environment for Spatial Reasoning. Under review. Website / Paper / “Can you untangle a knot in your head? Can a VLM?”
- Chen, Z., Gul, M. O., Chen, Y., Geng, G., Wu, A., & Artzi, Y. (2024). Retrospective Learning from Interactions. ACL 2025 Oral. Website / Paper / “Everything, in retrospect, is obvious.”
- Li, A. C., Chen, Z., Klassen, T. Q., Vaezipoor, P., Icarte, R. T., & McIlraith, S. A. (2024). Reward Machines for Deep RL in Noisy and Uncertain Environments. NeurIPS 2024 Poster. Paper / “Grounded reward specification is hard.”
- Chen, Z., Wang, H., Yeh, C. H., & Liu, X. (2022). Classify Respiratory Abnormality in Lung Sounds using STFT and a Fine-tuned ResNet18 Network. BioCAS 2022. Paper / “Will it cure my asthma?”
Miscellaneous
- I once put a battery in reverse and debugged for an hour (video). More stories like this.
- I programmed likely world’s least cost-efficient photocopier with a robot arm (video).
- I was certified for spending too much time on the Internet.