Research map / Image generation / Text-to-image diffusion
Deep Reward Supervisions for Tuning Text-to-Image Diffusion Models
Xiaoshi Wu, Yiming Hao, Manyuan Zhang, Keqiang Sun, Zhaoyang Huang, Guanglu Song, Yu Liu, Hongsheng Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| supervise early denoising steps with DRTune instead of only last steps | Stable Diffusion 1.5 · Symmetry reward | symmetry error: 0.7250 (DRaFT-10) -> 0.0551 | Improved |
| add stop gradient on denoising network input (ablation) | DRaFT-T · reward training convergence | convergence | Improved |
| add stop gradient on denoising network input (ablation) | AlignProp · reward training convergence | iterations to reach similar reward: roughly 50% of iterations | Improved |
| increase training steps K beyond 15 without stop gradient (ablation) | DRTune variants · gradient norm of LoRA parameters | gradient norm | Hurt |
| choose K = 0.1T training steps (ablation) | DRTune · T from 20 to 50 | final reward | Improved |
| choose early stop m = 0.4T (ablation) | DRTune · various T | final reward | Improved |
| use equally spaced training timesteps instead of random subset (ablation) | DRTune · reward training | convergence speed | Improved |
| fine-tune SDXL 1.0 with DRTune on HPS v2.1 | SDXL 1.0 · user study (400 prompts) | win rate vs SDXL: 69% win rate | Improved |
Closest papers by meaning
- DiffExp: Efficient Exploration in Reward Fine-tuning for Text-to-Image Diffusion Models (AAAI 2025 · similarity 0.90)
- Towards Better Alignment: Training Diffusion Models with Reinforcement Learning Against Sparse Rewards (CVPR 2025 · similarity 0.89)
- Directly Fine-Tuning Diffusion Models on Differentiable Rewards (ICLR 2024 · similarity 0.89)
- Using Human Feedback to Fine-tune Diffusion Models without Any Reward Model (CVPR 2024 · similarity 0.89)
- PRDP: Proximal Reward Difference Prediction for Large-Scale Reward Finetuning of Diffusion Models (CVPR 2024 · similarity 0.88)
- David and Goliath: Small One-step Model Beats Large Diffusion with Score Post-training (ICML 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
