VidLLMs CVPR2025

🚀New Paper & Dataset out! ImplicitQA: Going Beyond Frames Toward Implicit Video Reasoning 🎬 We challenge VideoQA models to reason about what’s not visible - like implied actions, hidden causality, and off-screen context. Paper: arxiv.org/abs/2506.21742 🧵👇 #VideoQA #AI #LLM

🎥 Missed the VideoLLMs Workshop at #CVPR2025? All keynote talks, oral presentations, challenge winner spotlights, and the panel discussion from the VideoLLMs Workshop are now available to watch online. 📺 Watch the full playlist here: youtube.com/playlist?list=…

Congrats to the winners of the Challenge Tracks at CVPR's VideoLLMs workshop ! Special thanks to Amazon Science and Apple for sponsoring awards worth $8,000 across 3 challenge tracks.

Congrats to the winners of Travel Grant awards from the VideoLLMs workshop at CVPR !

Post-Lunch session: Our distinguished panel consisting of Rene Vidal, Haoqi Fan, Salman Khan and moderated by Mubarak Shah is discussing the relative merits of VideoLLMs and Expert Models. Venue: Karl Dean Ballroom A1

A trip down memory lane in Prof Fahad Khan's talk on "Towards Detailed Video Understanding in Generative AI Era". Now ongoing at Karl Dean Ballroom A1 as part of the VideoLLMs workshop.

Dr Afshin Dehghan's talk is not beginning at the VideoLLMs workshop. Venue: Karl Dean Ballroom A1