Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Making Image Editing Easier via Adaptive Task Reformulation with Agentic Executions

About

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.

Bo Zhao, Kairui Guo, Runnan Du, Haiyang Sun, Pengshan Wang, Huan Yang, Kun Gai, Yixin Cao, Wei Ji• 2026

Related benchmarks

TaskDatasetResultRank
High-resolution perceptionHR-Bench-4K
Overall Score75.1
126
High-Resolution Visual PerceptionHRBench-8K
Overall Score72.6
29
Multimodal Perception and ReasoningMME-RealWorld-Lite
Overall Score53.2
21
Visual PerceptionV*
Overall Score90
11
Image EditingPICA
Average Score66.14
9
Image EditingRePlan
Quality4.12
9
Image EditingImgEdit Easy
Add Score4.47
9
Image EditingImgEdit Hard
Average Score4.19
9
Showing 8 of 8 rows

Other info

Follow for update