Add ReNeg: An end-toend method designed to learn improved Negative embeddings (CVPR 2025 Highlight)

[junpan19-new]

Model/Pipeline/Scheduler description

[ReNeg] is a reward-guided approach that directly learns Negative embeddings through gradient descent. The negative embedding learned within the same text embedding space exhibits strong generalization capabilities.

For example, using the same CLIP text encoder, the negative embedding learned on SD1.5 can be seamlessly transferred to text-to-image or even text-to-video models such as ControlNet, ZeroScope, and VideoCrafter2, resulting in consistent performance improvements across the board.

Open Source Status

• The model implementation is available.
• The model weights are available (Only relevant if addition is not a scheduler).

Provide useful links for the implementation

Official implementation: https://github.com/AMD-AIG-AIMA/ReNeg

[junpan19-new]

Hi,

The learned negative embeddings have been successfully converted to the safetensors format and uploaded to the Hub:

🔗 https://huggingface.co/XiaominDLUT/ReNeg

We appreciate your kind support and help!

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.