We consider the problem of Imitation Learning (IL) by actively querying noisy expert for feedback.

Learning modular object-centric representations is crucial for systematic generalization.

This work was done when Mingyu was at MIT.

The recent emergence of powerful Vision-Language models (VLMs) has significantly improved image captioning.

Understanding the intricate motivations behind human actions is paramount in developing sophisticated systems capable of nuanced behavior analysis.

Can a physicist make only a finite number of errors in the eternal quest to uncover the law of nature? This millennium-old philosophical problem, known as inductive inference, lies at the heart of epistemology.