DDA-Thinker: Decoupled Dual-Atomic Reinforcement Learning for Reasoning-Driven Image Editing

• URL: http://arxiv.org/abs/2604.25477v1
• Date: Tue, 28 Apr 2026 10:30:01 GMT
• Title: DDA-Thinker: Decoupled Dual-Atomic Reinforcement Learning for Reasoning-Driven Image Editing
• Abstract summary: We propose a Thinker-centric framework for the independent optimization of a planning module (Thinker) over a fixed generative model (Editor)<n>This framework decomposes feedback into two distinct atomic rewards implemented through verifiable checklists.<n>Experiments on reasoning-driven image editing benchmarks, including RISE-Bench and KRIS-Bench, demonstrate that our approach substantially improves overall performance.
• Score: 41.08605870394525
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent image editing models have achieved strong visual fidelity but often struggle with tasks requiring complex reasoning. To investigate and enhance the reasoning-grounded planning for image editing, we propose DDA-Thinker, a Thinker-centric framework designed for the independent optimization of a planning module (Thinker) over a fixed generative model (Editor). This decoupled Thinker-centric paradigm facilitates a controlled analysis of the planning module and makes its contribution under a fixed Editor easier to assess. To effectively guide this Thinker, we introduce a dual-atomic reinforcement learning framework. This framework decomposes feedback into two distinct atomic rewards implemented through verifiable checklists: a cognitive-atomic reward to directly assess the quality of the Thinker's executable plan, which serves as the actionable outcome of the Thinker's reasoning, and a visual-atomic reward to assess the final image quality. To improve checklist quality, our checklist synthesis is grounded not only in the source image and user instruction but also in a rational reference description of the ideal post-edit scene. To support this training, we further develop a two-stage data curation pipeline that first synthesizes a diverse and reasoning-focused dataset, then applies difficulty-aware refinement to curate an effective training curriculum for reinforcement learning. Extensive experiments on reasoning-driven image editing benchmarks, including RISE-Bench and KRIS-Bench, demonstrate that our approach substantially improves overall performance. Our method enables a community model to achieve results competitive with strong proprietary models, highlighting the practical potential of Thinker-centric optimization under a fixed-editor setting.

Related papers

• Unlocking the Potential of Image Editing via Concept Scaling and Dense Supervision [38.31785674228309]
  We establish a comprehensive hierarchical taxonomy featuring over 1,000 fine-grained edit concepts.<n>We build ConceptEdit-12M, a massive dataset of 12 million high-quality editing pairs.<n>We propose a dense supervision training strategy that synthesizes multiple non-interfering concepts into single image pairs.
  arXiv  Detail & Related papers  (2026-08-17T17:04:25Z)
• Edit-Compass & EditReward-Compass: A Unified Benchmark for Image Editing and Reward Modeling [18.3502725927898]
  We introduce Edit- and Edit-Reward, a unified evaluation suite for image editing and reward modeling.<n> Edit-Reward includes 2,388 annotated instances spanning six progressively challenging task categories.<n>It adopts a fine-grained multidimensional evaluation framework based on structured reasoning and carefully designed rubrics.
  arXiv  Detail & Related papers  (2026-05-13T06:33:54Z)
• ReasonEdit: Towards Interpretable Image Editing Evaluation via Reinforcement Learning [86.61218827780675]
  We present ReasonEdit, an evaluation tool for text-guided image editing.<n>It is trained using reward signals derived from RE-Reward and the Group Relative Policy Optimization (GRPO) algorithm.<n>It is capable of generating high-quality interpretable evaluation text, enabling more transparent and trustworthy assessment for image editing.
  arXiv  Detail & Related papers  (2026-05-08T09:23:26Z)
• Q-DeepSight: Incentivizing Thinking with Images for Image Quality Assessment and Refinement [58.15004031934379]
  We propose Q-DeepSight, a think-with-image framework that emulates this human-like process.<n>Q-DeepSight achieves state-of-the-art performance across diverse benchmarks, including natural, restored, and AI-generated content.<n>We demonstrate its practical value with Perceptual-in-Generation (PiG), a training-free framework where Q-DeepSight's diagnoses guide iterative image enhancement.
  arXiv  Detail & Related papers  (2026-04-18T06:10:57Z)
• Unsupervised Synthetic Image Attribution: Alignment and Disentanglement [55.853285140682665]
  We propose a simple yet effective unsupervised method called Alignment and Disentanglement.<n>Specifically, we begin by performing basic concept alignment using contrastive self-supervised learning.<n>Next, we enhance the model's attribution ability by promoting representation disentanglement with the Infomax loss.
  arXiv  Detail & Related papers  (2026-01-30T07:31:53Z)
• ThinkRL-Edit: Thinking in Reinforcement Learning for Reasoning-Centric Image Editing [33.888289858260706]
  Reinforcement learning (RL) has been investigated for improving the quality of image editing.<n>RL faces three key challenges: (1) limited reasoning exploration confined to denoising, (2) biased reward fusion, and (3) unstable VLM-based instruction rewards.<n>We propose ThinkRL-Edit, a reasoning-centric RL framework that decouples visual reasoning from image synthesis.
  arXiv  Detail & Related papers  (2026-01-06T23:43:00Z)
• EditThinker: Unlocking Iterative Reasoning for Any Image Editor [72.28251670314451]
  We propose a deliberative editing framework to 'think' while they edit.<n>We train a single MLLM, EditThinker, to act as the reasoning engine of this framework.<n>We employ reinforcement learning to align the EditThinker's thinking with its editing, thereby generating more targeted instruction improvements.
  arXiv  Detail & Related papers  (2025-12-05T18:58:09Z)
• EditScore: Unlocking Online RL for Image Editing via High-Fidelity Reward Modeling [71.8265422228785]
  Reinforcement Learning (RL) offers a promising solution, but its adoption in image editing has been hindered by the lack of a high-fidelity, efficient reward signal.<n>We present a comprehensive methodology to overcome this barrier, centered on the development of a state-of-the-art, specialized reward model.
  arXiv  Detail & Related papers  (2025-09-28T14:28:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.