|
Max Wilcoxson I am a PhD student at MIT CSAIL advised by Vincent Sitzmann. I did my B.A. at Berkeley where I was advised by Sergey Levine. During my undergrad, I received the Warren Y. Dere Design Award. |
|
ResearchI am interested in building embodied agents that improve themselves, combining video generation models, reinforcement learning, and exploration. Some papers are highlighted. |
|
Self-Improving Video-to-Action Models
Hyunwoo Ryu*, Eric Chen*, Max Wilcoxson*, Vincent Sitzmann Under review project page Mirror solves embodied tasks without expert action labels or task rewards through repeated self-play with a large-scale video planner. |
|
|
Leveraging Skills from Unlabeled Prior Data for Efficient Online Exploration
Max Wilcoxson*, Qiyang Li*, Kevin Frans, Sergey Levine ICML, 2025 project page / arXiv / bibtex Leveraging unlabeled offline data for both skill pretraining and optimistic relabelling leads to extremely efficient online exploration. |
|
Polynomial Regression as a Task for Understanding In-context Learning Through Finetuning and Alignment
Max Wilcoxson*, Morten Svendgård*, Ria Doshi*, Dylan Davis*, Reya Vir, Anant Sahai ICML Workshop on In-Context Learning, 2024 arxiv / bibtex Univariate polynomial regression is a simple task which captures in-context fine-tuning and alignment behavior, facilitating informative visualizations and deeper understanding. |
|
|