image-rm-sd3.5-medium-pickapicv2

Selected RM Checkpoint

This is the selected Image RM checkpoint trained on liuhuohuo2/pick-a-pic-v2, using stabilityai/stable-diffusion-3.5-medium as the backbone.

Configuration

• Checkpoint: checkpoint-final
• T5 maximum sequence length: 256
• Learning rate: 3e-5
• Guidance scale: 5
• Effective batch size: 128
• Resolution: 512
• Seed: 42

Final Validation Metrics

eval was used as the primary selection metric and eval_unique as the complementary metric. This run provided the best overall balance among the completed T5-256 RM runs, with only mild late drift in eval loss and no late rise in eval_unique loss.

Use this checkpoint for reward labeling and reward-based alignment.