# Diagnose-to-Generate: Eliminating Structural Distortions in Flow Models via Patch-Level Grounded Alignment

arXiv: https://arxiv.org/abs/2610.17882 · HTML: https://mlsanity.com/paper/2610.17882/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Diffusion and multimodal model alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 31 of its 51 references that are in the map.
- First posted on arXiv 13 Oct 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Diffusion and multimodal model alignment](https://mlsanity.com/alignment/diffusion-and-multimodal-model-alignment/)

## Builds on (most cited first, 30 of 31)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://mlsanity.com/paper/2412.03603/) (2024)
- [InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency](https://mlsanity.com/paper/2508.18265/) (2025)
- [Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression](https://mlsanity.com/paper/1902.09630/) (2019)
- [ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2304.05977/) (2023)
- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2305.01569/) (2023)
- [Emerging Properties in Unified Multimodal Pretraining](https://mlsanity.com/paper/2505.14683/) (2025)
- [Flow-GRPO: Training Flow Matching Models via Online RL](https://mlsanity.com/paper/2505.05470/) (2025)
- [GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning](https://mlsanity.com/paper/2507.01006/) (2025)
- [DanceGRPO: Unleashing GRPO on Visual Generation](https://mlsanity.com/paper/2505.07818/) (2025)
- [Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer](https://mlsanity.com/paper/2511.22699/) (2025)
- [Seedance 1.0: Exploring the Boundaries of Video Generation Models](https://mlsanity.com/paper/2506.09113/) (2025)
- [T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT](https://mlsanity.com/paper/2505.00703/) (2025)
- [Q-Insight: Understanding Image Quality via Visual Reinforcement Learning](https://mlsanity.com/paper/2503.22679/) (2025)
- [LongCat-Video Technical Report](https://mlsanity.com/paper/2510.22200/) (2025)
- [HumanRefiner: Benchmarking Abnormal Human Generation and Refining with Coarse-to-fine Pose-Reversible Guidance](https://mlsanity.com/paper/2407.06937/) (2024)
- [GIE-Bench: Towards Grounded Evaluation for Text-Guided Image Editing](https://mlsanity.com/paper/2505.11493/) (2025)
- [HEIE: MLLM-Based Hierarchical Explainable AIGC Image Implausibility Evaluator](https://mlsanity.com/paper/2411.17261/) (2024)
- [RealGen: Photorealistic Text-to-Image Generation via Detector-Guided Rewards](https://mlsanity.com/paper/2512.00473/) (2025)
- [RePlan: Reasoning-guided Region Planning for Complex Instruction-based Image Editing](https://mlsanity.com/paper/2512.16864/) (2025)
- [Re-Align: Structured Reasoning-guided Alignment for In-Context Image Generation and Editing](https://mlsanity.com/paper/2601.05124/) (2026)
- [Evaluating and Predicting Distorted Human Body Parts for Generated Images](https://mlsanity.com/paper/2503.00811/) (2025)
- [Thinking with Frames: Generative Video Distortion Evaluation via Frame Reward Model](https://mlsanity.com/paper/2601.04033/) (2026)

## Abstract (arXiv)

Flow-based image generators achieve strong semantic synthesis but still produce localized structural distortions, such as malformed hands and limbs. Existing post-training methods typically optimize global scalar rewards, which provide weak spatial credit assignment and may alter well-formed regions while correcting local defects. We present textitDiagnose-to-Generate (D2G), a grounded alignment framework that separates visual diagnosis from generative optimization. We construct Achilles-600K, a region-annotated dataset for structural anomalies, and train ARGUS through supervised cold-start and Region-Aware Group Relative Policy Optimization to predict anomaly categories and bounding boxes. These diagnoses are converted into confidence-weighted spatial rewards for our Patch-level Flow-GRPO. We further identify that global stochastic exploration can shift semantic regions across group rollouts, making patch-wise relative advantages spatially inconsistent. Our Trajectory-Preserving Stochastic Sampling (TPSS) anchors early denoising with deterministic ODE trajectories and introduces stochasticity only at later steps, thereby limiting intra-group spatial drift while retaining local exploration. Experiments on FLUX.1-dev and SD3.5 show improved structural fidelity with preserved FID and CLIP scores. Consistent gains under HumanRefiner, zero-shot GPT-4.1, and blinded human evaluation further demonstrate that the improvements extend beyond the training reward model.