Reward Model training data with FLUX.1-Kontext-dev edited negative samples.
📊 Data Contents (~173GB total)
🔧 Edit Operations (Negative Sample Generation)
content_remove
Remove objects… See the full description on the dataset page:
https://huggingface.co/datasets/mohan2/StyleTransfer-Reward-FluxEdit.
