Abstract
Tool-based image editing (image retouching) is commonly formulated with autoregressive multimodal large language models (MLLMs) that sequentially generate reasoning, tool selections, and parameter values. In this work, we present a novel approach to tool-based image editing by framing the task as a flow matching problem. We introduce FlowTool, a framework that directly models the distribution of high-quality tool parameters conditioned on the input image and user instruction using conditional rectified flow. FlowTool combines a vision-language model backbone for multimodal understanding with a Diffusion Transformer parameter generator that transforms Gaussian noise into an editing plan. We train FlowTool with a two-stage supervised flow-matching curriculum, followed by reward-based post-training. Across MMArt-Bench, FlowTool-Eval, ArtEdit-Bench, and MIT-Adobe5K, FlowTool achieves significantly stronger reference-based performance than specialized MLLM editing agents and proprietary MLLMs, while remaining competitive with proprietary models under reference-free evaluation. Moreover, FlowTool significantly improves inference efficiency, reducing latency by at least $50\times$ while requiring nearly $2\times$ less memory than the compared baselines. These results demonstrate that tool-based image editing can be effectively modeled as conditional generation over structured continuous editing parameters, without autoregressive reasoning.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Le, T. L. V., Walton, S., Yoon, S., Kveton, B., Bui, T., Yang, E., & Lai, V. (2026). FlowTool: Controlling Tool Parameter in Image Retouching via Flow Matching. https://omanscience.com/en/articles/flowtool-controlling-tool-parameter-in-image-retouching-via-flow-matching
MLA 9
Le, Thanh-Long V., et al. "FlowTool: Controlling Tool Parameter in Image Retouching via Flow Matching." https://omanscience.com/en/articles/flowtool-controlling-tool-parameter-in-image-retouching-via-flow-matching.
Chicago (author–date)
Le, Thanh-Long V., Steven Walton, Seunghyun Yoon, Branislav Kveton, Trung Bui, Eunho Yang, and Viet Lai. 2026. "FlowTool: Controlling Tool Parameter in Image Retouching via Flow Matching." https://omanscience.com/en/articles/flowtool-controlling-tool-parameter-in-image-retouching-via-flow-matching.
Harvard
Le, T. L. V., Walton, S., Yoon, S., Kveton, B., Bui, T., Yang, E. and Lai, V. (2026) 'FlowTool: Controlling Tool Parameter in Image Retouching via Flow Matching', Available at: https://omanscience.com/en/articles/flowtool-controlling-tool-parameter-in-image-retouching-via-flow-matching.
Vancouver
Le TLV, Walton S, Yoon S, Kveton B, Bui T, Yang E, et al. FlowTool: Controlling Tool Parameter in Image Retouching via Flow Matching. https://omanscience.com/en/articles/flowtool-controlling-tool-parameter-in-image-retouching-via-flow-matching
IEEE
T. L. V. Le, S. Walton, S. Yoon, B. Kveton, T. Bui, E. Yang, and V. Lai, "FlowTool: Controlling Tool Parameter in Image Retouching via Flow Matching," https://omanscience.com/en/articles/flowtool-controlling-tool-parameter-in-image-retouching-via-flow-matching.