view post Post 1012 📣 Happy to introduce, CinePile, a long video QA dataset and benchmark! 300k train and 5k test split. The dataset is live on HF right now! :) Link: tomg-group-umd/cinepilePls check it out and let us know if you have any comments or suggestions! :)
VLM-hallu Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding Paper • 2311.16922 • Published Nov 28, 2023 • 1 RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback Paper • 2312.00849 • Published Dec 1, 2023 • 8
Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding Paper • 2311.16922 • Published Nov 28, 2023 • 1
RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback Paper • 2312.00849 • Published Dec 1, 2023 • 8