https://github.com/mihirp1998/VADER

Video Diffusion Alignment via Reward Gradients. We improve a variety of video diffusion models such as VideoCrafter, OpenSora, ModelScope and StableVideoDiffusion by finetuning them using various reward models such as HPS, PickScore, VideoMAE, VJEPA, YOLO, Aesthetics etc.
https://github.com/mihirp1998/VADER

alignment diffusion reinforcement-learning reinforcement-learning-human-feedback rl rlhf vader video-diffusion video-diffusion-alignment

Last synced: about 1 month ago
JSON representation

Video Diffusion Alignment via Reward Gradients. We improve a variety of video diffusion models such as VideoCrafter, OpenSora, ModelScope and StableVideoDiffusion by finetuning them using various reward models such as HPS, PickScore, VideoMAE, VJEPA, YOLO, Aesthetics etc.

• Host: GitHub
• URL: https://github.com/mihirp1998/VADER
• Owner: mihirp1998
• Created: 2024-06-23T21:44:57.000Z (3 months ago)
• Default Branch: master
• Last Pushed: 2024-07-22T19:31:36.000Z (about 2 months ago)
• Last Synced: 2024-07-23T22:53:36.101Z (about 2 months ago)
• Topics: alignment, diffusion, reinforcement-learning, reinforcement-learning-human-feedback, rl, rlhf, vader, video-diffusion, video-diffusion-alignment
• Language: Python
• Homepage: https://vader-vid.github.io/
• Size: 163 MB
• Stars: 127
• Watchers: 6
• Forks: 6
• Open Issues: 4
• Metadata Files:
  • Readme: readme.md

Awesome Lists containing this project

• awesome-diffusion-categorized - [Code