Jaemin Cho

@jmin__cho

Assistant Prof @JHUCompSci | Prev @Allen_AI PhD @UNCCS | multimodal and embodied AI

Its creators say the benchmark pairs simple goals with challenging physics to test both models beyond simple robot-control tasks.

5 Sources, 12d ago, updated 12d ago

A user wants to create a dance transcription system akin to music transcription. For now, they say they do the work by hand.

2 Sources, 24d ago, updated 24d ago

Incoming Johns Hopkins professor shares details on AnchorWeave video method.

1 Source, 44d ago, first seen 44d ago

AI researcher Jaemin Cho references new work on multimodal models via X.

1 Source, 44d ago, first seen 44d ago

AI researcher Jaemin Cho shares work on visual representation alignment via co-denoising.

2 Sources, 44d ago, first seen 44d ago

Incoming Johns Hopkins professor shares plans to keynote two workshops during the September conference.

3 Sources, 44d ago, first seen 44d ago

Featured in