Building a Precise Video Language with Human-AI Oversight
Paper ā¢ 2604.21718 ā¢ Published ā¢ 17
Zhiqiu Lin PRO
zhiqiulin
AI & ML interests
None yet
Recent Activity
posted an update 25 days ago
š VQAScore now supports text-to-video evaluation!
VQAScore scores how well a generated image or video matches a prompt by asking a VLM "does this show {prompt}?" and using P(Yes). It became a go-to evaluation metric and reward model for image generation (2M+ downloads), and we just added text-to-video support across 20+ VLMs (GPT, Gemini, Qwen). Free and open-source, and it keeps improving as VLMs improve.
š» Code: https://github.com/linzhiqiu/t2v_metrics
š Paper: https://arxiv.org/abs/2404.01291
š§µ Launch thread + demo video: https://x.com/ZhiqiuLin/status/2064316582461841499 updated a dataset 30 days ago
zhiqiulin/caption_export liked a dataset 2 months ago
chancharikm/CHAI_testsetOrganizations
upvoted a paper 2 months ago
upvoted 3
papers about 1 year ago
Sparse Attention Vectors: Generative Multimodal Model Features Are Discriminative Vision-Language Classifiers
Paper ā¢ 2412.00142 ā¢ Published ā¢ 5
Towards Understanding Camera Motions in Any Video
Paper ā¢ 2504.15376 ā¢ Published ā¢ 157
AerialMegaDepth: Learning Aerial-Ground Reconstruction and View Synthesis
Paper ā¢ 2504.13157 ā¢ Published ā¢ 20
upvoted 2
papers over 1 year ago
upvoted a paper almost 2 years ago
upvoted a paper about 2 years ago