# RetouchIQ: MLLM Agents for Instruction-Based Image Retouching with Generalist Reward

arXiv: https://arxiv.org/abs/2602.17558 · HTML: https://mlsanity.com/paper/2602.17558/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Image editing and quality assessment, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 30 of its 41 references that are in the map.
- First posted on arXiv 19 Feb 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Image editing and quality assessment](https://mlsanity.com/mllm/image-editing-and-quality-assessment/)
- LLM agents and tool use / [Multimodal and embodied agents](https://mlsanity.com/agents/multimodal-and-embodied-agents/)
- 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 30)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning](https://mlsanity.com/paper/2507.01006/) (2025)
- [VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning](https://mlsanity.com/paper/2504.08837/) (2025)
- [GUI-R1 : A Generalist R1-Style Vision-Language Action Model For GUI Agents](https://mlsanity.com/paper/2504.10458/) (2025)
- [Inference-Time Scaling for Generalist Reward Modeling](https://mlsanity.com/paper/2504.02495/) (2025)
- [Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs](https://mlsanity.com/paper/2401.11708/) (2024)
- [UI-R1: Enhancing Efficient Action Prediction of GUI Agents by Reinforcement Learning](https://mlsanity.com/paper/2503.21620/) (2025)
- [GenArtist: Multimodal LLM as an Agent for Unified Image Generation and Editing](https://mlsanity.com/paper/2407.05600/) (2024)
- [Uncovering the Disentanglement Capability in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2212.08698/) (2022)
- [VRAG-RL: Empower Vision-Perception-Based RAG for Visually Rich Information Understanding via Iterative Reasoning with Reinforcement Learning](https://mlsanity.com/paper/2505.22019/) (2025)
- [Harmonizer: Learning to Perform White-Box Image and Video Harmonization](https://mlsanity.com/paper/2207.01322/) (2022)
- [SophiaVL-R1: Reinforcing MLLMs Reasoning with Thinking Reward](https://mlsanity.com/paper/2505.17018/) (2025)
- [RestoreAgent: Autonomous Image Restoration Agent via Multimodal Large Language Models](https://mlsanity.com/paper/2407.18035/) (2024)
- [JarvisArt: Liberating Human Artistic Creativity via an Intelligent Photo Retouching Agent](https://mlsanity.com/paper/2506.17612/) (2025)
- [Unpaired Image Enhancement Featuring Reinforcement-Learning-Controlled Image Editing Software](https://mlsanity.com/paper/1912.07833/) (2019)
- [We-Math 2.0: A Versatile MathBook System for Incentivizing Visual Mathematical Reasoning](https://mlsanity.com/paper/2508.10433/) (2025)
- [Learning by Planning: Language-Guided Global Image Editing](https://mlsanity.com/paper/2106.13156/) (2021)
- [GenMAC: Compositional Text-to-Video Generation with Multi-Agent Collaboration](https://mlsanity.com/paper/2412.04440/) (2024)
- [Skywork-R1V3 Technical Report](https://mlsanity.com/paper/2507.06167/) (2025)
- [RSFNet: A White-Box Image Retouching Approach using Region-Specific Color Filters](https://mlsanity.com/paper/2303.08682/) (2023)
- [DiffRetouch: Using Diffusion to Retouch on the Shoulder of Experts](https://mlsanity.com/paper/2407.03757/) (2024)
- [Generative RLHF-V: Learning Principles from Multi-modal Human Preference](https://mlsanity.com/paper/2505.18531/) (2025)
- [MonetGPT: Solving Puzzles Enhances MLLMs' Image Retouching Skills](https://mlsanity.com/paper/2505.06176/) (2025)
- [Training-Free Large Model Priors for Multiple-in-One Image Restoration](https://mlsanity.com/paper/2407.13181/) (2024)
- [MoFRR: Mixture of Diffusion Models for Face Retouching Restoration](https://mlsanity.com/paper/2507.19770/) (2025)
- [RetouchLLM: Training-free Code-based Image Retouching with Vision Language Models](https://mlsanity.com/paper/2510.08054/) (2025)

## Built on by (most cited first, 4 of 4)

- [Agentic Visual Generation: From Generative Models to Agentic Control](https://mlsanity.com/paper/2609.06758/) (2026)
- [SmartPhotoCrafter: Unified Reasoning, Generation and Optimization for Automatic Photographic Image Editing](https://mlsanity.com/paper/2604.19587/) (2026)
- [Dotting the Eye: An Intent-Driven Image Retouching Agent for Visual Focus Enhancement](https://mlsanity.com/paper/2609.01148/) (2026)
- [FlowTool: Controlling Tool Parameter in Image Retouching via Flow Matching](https://mlsanity.com/paper/2609.35673/) (2026)

## Abstract (arXiv)

Recent advances in multimodal large language models (MLLMs) have shown great potential for extending vision-language reasoning to professional tool-based image editing, enabling intuitive and creative editing. A promising direction is to use reinforcement learning (RL) to enable MLLMs to reason about and execute optimal tool-use plans within professional image-editing software. However, training remains challenging due to the lack of reliable, verifiable reward signals that can reflect the inherently subjective nature of creative editing. In this work, we introduce RetouchIQ, a framework that performs instruction-based executable image editing through MLLM agents guided by a generalist reward model. RetouchIQ interprets user-specified editing intentions and generates corresponding, executable image adjustments, bridging high-level aesthetic goals with precise parameter control. To move beyond conventional, rule-based rewards that compute similarity against a fixed reference image using handcrafted metrics, we propose a generalist reward model, an RL fine-tuned MLLM that evaluates edited results through a set of generated metrics on a case-by-case basis. Then, the reward model provides scalar feedback through multimodal reasoning, enabling reinforcement learning with high-quality, instruction-consistent gradients. We curate an extended dataset with 190k instruction-reasoning pairs and establish a new benchmark for instruction-based image editing. Experiments show that RetouchIQ substantially improves both semantic consistency and perceptual quality over previous MLLM-based and diffusion-based editing systems. Our findings demonstrate the potential of generalist reward-driven MLLM agents as flexible, explainable, and executable assistants for professional image editing.