Research map / Image generation / Text-to-image diffusion
Self-Corrected Image Generation with Explainable Latent Rewards
Yinyi Luo, Hrishikesh Gokhale, Marios Savvides, Jindong Wang, Shengfeng He
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 |
|---|---|---|---|
| add RL objective (CLIPScore reward) (ablation) | xLARD · GenEval | overall (%): 77.68 -> 78.78 | Improved |
| swap CLIPScore reward for Sentence-BERT reward (ablation) | xLARD · GenEval | overall (%): 78.78 -> 78.42 | No clear effect |
| add confidence-guided latent modulation (ablation) | xLARD · GenEval | overall: -3.3% GenEval drop when removed | Improved |
| add latent anchor (ablation) | xLARD · GenEval | overall (%): 76.90 (without anchor) | Improved |
| add counting reward (ablation) | xLARD · GenEval | overall (%): 76.45 -> 81.24 | Improved |
| add color reward (ablation) | xLARD · GenEval | overall (%): 78.92 -> 81.24 | Improved |
| add position reward (ablation) | xLARD · GenEval | overall (%): 79.15 -> 81.24 | Improved |
| add counting reward (ablation) | xLARD · DPG-Bench | overall (%): 83.10 -> 86.72 | Improved |
| add color reward (ablation) | xLARD · DPG-Bench | overall (%): 84.55 -> 86.72 | Improved |
| add xLARD corrector to OmniGen2 | OmniGen2 · GenEval | overall (%): 77.03 -> 81.29 | Improved |
| add xLARD corrector to BAGEL | BAGEL · GenEval | overall (%): 78.97 -> 81.38 | Improved |
| add xLARD corrector to Show-O | Show-O · GenEval | overall (%): 68.05 -> 74.89 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- OmniGen — OmniGen: Unified Image Generation (CVPR 2025) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Learning to Generate via Understanding: Understanding-Driven Intrinsic Rewarding for Unified Multimodal Models (CVPR 2026 · similarity 0.89)
- Progressive Text-to-Image Diffusion with Soft Latent Direction (AAAI 2024 · similarity 0.88)
- PromptCoT: Align Prompt Distribution via Adapted Chain-of-Thought (CVPR 2024 · similarity 0.88)
- Janus-Pro-R1: Advancing Collaborative Visual Comprehension and Generation via Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- LMFusion: Adapting Pretrained Language Models for Multimodal Generation (NeurIPS 2025 · similarity 0.88)
- LLM Blueprint: Enabling Text-to-Image Generation with Complex and Detailed Prompts (ICLR 2024 · 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.
