EditRefiner: A Human-Aligned Agentic Framework for Image Editing Refinement

• URL: http://arxiv.org/abs/2605.07457v1
• Date: Fri, 08 May 2026 09:05:08 GMT
• Title: EditRefiner: A Human-Aligned Agentic Framework for Image Editing Refinement
• Abstract summary: EditRefiner is an agentic framework that reformulates post-editing correction as a human-like perception-reasoning-action-evaluation loop.<n>It consistently outperforms state-of-the-art methods in distortion, diagnose accuracy and human perception alignment.
• Score: 76.76247443244293
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent text-guided image editing (TIE) models have made remarkable progress, yet edited images still frequently suffer from fine-grained issues such as unnatural objects, lighting mismatch, and unexpected changes. Existing refinement approaches either rely on costly iterative regeneration or employ vision-language models (VLMs) with weak spatial grounding, often resulting in semantic drift and unreliable local corrections. To address these limitations, we first construct EditFHF-15K, a dataset of fine-grained human feedback for edited images, comprising (1) 15K images from 12 TIE models spanning 43 editing tasks, (2) 60K annotated artifact regions and 80K editing failure regions, each accompanied by textual reasoning, and (3) 45K mean opinion scores (MOSs) assessing perceptual quality, instruction following, and visual consistency. Based on EditFHF-15K, we propose EditRefiner, a hierarchical, interpretable, and human-aligned agentic framework that reformulates post-editing correction as a human-like perception-reasoning-action-evaluation loop. Specifically, we introduce: (1) a perception agent that detects contextual saliency maps of artifacts and editing failures, (2) a reasoning agent that interprets these perceptual cues to perform human-aligned diagnostic inference, (3) an action agent that uses the reasoning output to plan and execute localized re-editing, and (4) an evaluation agent that assesses the re-edited image and guides the action agent on whether further refinements are required. Extensive experiments demonstrate that EditRefiner consistently outperforms state-of-the-art methods in distortion localization, diagnose accuracy and human perception alignment, establishing a new paradigm for self-corrective and perceptually reliable image editing. The code is available at https://github.com/IntMeGroup/EditRefiner.

Related papers

• 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)
• Is This Edit Correct? A Multi-Dimensional Benchmark for Reasoning-Aware Image Editing [45.07920210103563]
  Diffusion-based image editing has achieved strong visual fidelity under natural language instructions.<n>Most existing systems still operate at the level of surface instruction following, without reasoning about the implicit contextual constraints.<n>We introduce RE-Edit, a benchmark for REasoning-aware image editing.
  arXiv  Detail & Related papers  (2026-04-16T08:05:42Z)
• CAMEO: A Conditional and Quality-Aware Multi-Agent Image Editing Orchestrator [8.901958956012072]
  Conditional image editing is crucial in scenarios requiring strict structural control.<n>Most approaches rely on single-step generation.<n>We propose textbfCAMEO, a structured multi-agent framework that reformulates conditional editing as a quality-aware, feedback-driven process.
  arXiv  Detail & Related papers  (2026-04-03T16:27:02Z)
• Evaluating Image Editing with LLMs: A Comprehensive Benchmark and Intermediate-Layer Probing Approach [48.01137214342501]
  We introduce TIEdit, a benchmark for systematic evaluation of text-guided image editing methods.<n>To obtain reliable subjective ratings, 20 experts are recruited to produce 307,200 raw subjective ratings.<n>We further propose EditProbe, an LLM-based evaluator that estimates editing quality via intermediate-layer probing of hidden representations.
  arXiv  Detail & Related papers  (2026-03-20T09:05:10Z)
• Agentic Retoucher for Text-To-Image Generation [48.80766311858762]
  Agentic Retoucher is a hierarchical decision-driven framework that reformulates post-generation correction as a human-like perception-reasoning-action loop.<n>This design integrates perceptual evidence, linguistic reasoning, and controllable correction into a unified, self-corrective decision process.<n>Experiments demonstrate that Agentic Retoucher consistently outperforms state-of-the-art methods in perceptual quality, distortion localization and human preference alignment.
  arXiv  Detail & Related papers  (2026-01-05T12:06:43Z)
• Charts Are Not Images: On the Challenges of Scientific Chart Editing [66.38730113476677]
  textitFigEdit is a benchmark for scientific figure editing comprising over 30,000 samples.<n>Our benchmark demonstrates the profound limitations of pixel-level manipulation.<n>By releasing textitFigEdit, we aim to enable systematic progress in structure-aware figure editing.
  arXiv  Detail & Related papers  (2025-11-30T06:13:48Z)
• EditInspector: A Benchmark for Evaluation of Text-Guided Image Edits [22.762414256693265]
  We introduce EditInspector, a novel benchmark for evaluation of text-guided image edits.<n>We leverage EditInspector to evaluate the performance of state-of-the-art (SoTA) vision and language models in assessing edits.<n>Our findings indicate that current models struggle to evaluate edits comprehensively and frequently hallucinate when describing the changes.
  arXiv  Detail & Related papers  (2025-06-11T17:58:25Z)
• GIE-Bench: Towards Grounded Evaluation for Text-Guided Image Editing [60.66800567924348]
  We introduce a new benchmark designed to evaluate text-guided image editing models.<n>The benchmark includes over 1000 high-quality editing examples across 20 diverse content categories.<n>We conduct a large-scale study comparing GPT-Image-1 against several state-of-the-art editing models.
  arXiv  Detail & Related papers  (2025-05-16T17:55:54Z)

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.