Making Image Editing Easier via Adaptive Task Reformulation with Agentic Executions

TL;DR

This paper introduces an adaptive task reformulation framework that enhances instruction-guided image editing by dynamically transforming tasks, leading to improved performance especially on challenging cases without changing the underlying models.

Contribution

It proposes a novel framework that reformulates image editing tasks using a multi-modal language model agent, significantly improving results across multiple benchmarks and editing backbones.

Findings

Consistent performance improvements on ImgEdit, PICA, and RePlan benchmarks.

Large gains observed on challenging editing cases.

Task reformulation is a key factor in improving image editing outcomes.

Abstract

Instruction guided image editing has advanced substantially with recent generative models, yet it still fails to produce reliable results across many seemingly simple cases. We observe that a large portion of these failures stem not from insufficient model capacity, but from poorly formulated editing tasks, such as those involving small targets, implicit spatial relations, or under-specified instructions. In this work, we frame image editing failures as a task formulation problem and propose an adaptive task reformulation framework that improves editing performance without modifying the underlying model. Our key idea is to transform the original image-instruction pair into a sequence of operations that are dynamically determined and executed by a MLLM agent through analysis, routing, reformulation, and feedback-driven refinement. Experiments on multiple benchmarks, including ImgEdit, PICA, and RePlan, across diverse editing backbones such as Qwen Image Edit and Nano Banana, show consistent improvements, with especially large gains on challenging cases. These results suggest that task reformulation is a critical but underexplored factor, and that substantial gains can be achieved by better matching editing tasks to the effective operating regime of existing models.