Apply for a GPU community grant: Academic project

#1
by sming256 - opened

https://ivul-kaust.github.io/projects/videoauto-r1/

We propose VideoAuto-R1, a video understanding framework that adopts a "reason-when-necessary" strategy. During training, our approach follows a Thinking Once, Answering Twice paradigm: the model first generates an initial answer, then performs reasoning, and finally outputs a reviewed answer. Both answers are supervised via verifiable rewards. During inference, the model uses the confidence score of the initial answer to determine whether to proceed with reasoning.

Our models and demo are hosted on HuggingFace. We kindly request additional GPU resources to ensure a smooth and responsive demo experience for users.

Sign up or log in to comment