CoRL 2025 - 4 Papers Accepted (including one for Oral presentation)!!

The lab has been extremely fortunate this year, with 4 papers accepted to the prestigious Conference on Robot Learning (CORL 2025), which will take place 27-30 September 2025 in Seoul, Korea. Two of these papers are the result of international collaborations, borne from student visits at other universities.

The preprints of all papers will be available soon from OpenReview and/or ArXiv (links from the publication page), and for some of them you may also find a link to the project’s webpage and huggingface demo. The links below will be updated shortly, as these are still intermediate versions (and the openreview link for one of them isn’t public yet).

• Search-TTA: A Multimodal Test-Time Adaptation Framework for Visual Search in the Wild (Collaboration with Prof. Florian Shkurti at UofT)
• CogniPlan: Uncertainty-Guided Path Planning with Conditional Generative Layout Prediction
• Multimodal Fused Learning for Solving the Generalized Traveling Salesman Problem in Robotic Task Planning
• (ORAL) Latent Theory of Mind: A Decentralized Diffusion Architecture for Cooperative Manipulation (Collaboration with Prof. Mac Schwager at Stanford, link to paper coming soon)