arXiv 2603.001412026-03-26cs.CVcs.AIcs.LGeess.IV

从尺度到速度：面向图像编辑的自适应测试时间缩放

From Scale to Speed: Adaptive Test-Time Scaling for Image Editing

Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）；AMAP, Alibaba Group（阿里巴巴集团高德地图）；University of Queensland（昆士兰大学）

Xiangyan Qu, Zhenlong Yuan, Jing Tang, Rui Chen, Datao Tang, Meng Yu, Lei Sun, Yancheng Bai, Xiangxiang Chu, Gaopeng Gou, Gang Xiong, Yujun Cai

备注：: Accepted to the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出ADE-CoT框架，通过动态资源分配、编辑特定验证和深度优先停止策略，提升图像编辑效率与性能，实验显示在同等采样预算下性能优于现有方法。

摘要：

图像链式思维（Image-CoT）是一种通过延长推理时间来提升图像生成的测试时间缩放范式。大多数Image-CoT方法专注于文本到图像（T2I）生成。不同于T2I生成，图像编辑是目标导向的：解决方案空间由源图像和指令约束。这种不匹配导致在应用Image-CoT到编辑时面临三个挑战：固定采样预算下的资源分配低效、通用MLLM分数在早期验证中不可靠、以及大规模采样导致的冗余编辑结果。为了解决这些问题，我们提出了ADaptive Edit-CoT（ADE-CoT），一种按需测试时间缩放框架，以增强编辑效率和性能。它结合了三个关键策略：（1）难度感知的资源分配，根据估计的编辑难度动态分配预算；（2）早期剪枝中的编辑特定验证，利用区域定位和标题一致性选择有前途的候选者；（3）深度优先机会性停止，由实例特定验证器引导，当找到意图一致的结果时终止。在三个最先进编辑模型（Step1X-Edit、BAGEL、FLUX.1 Kontext）上进行的大量实验显示，ADE-CoT在性能-效率权衡上表现优异。在可比的采样预算下，ADE-CoT在性能上优于Best-of-N超过2倍。

英文摘要：

Image Chain-of-Thought (Image-CoT) is a test-time scaling paradigm that improves image generation by extending inference time. Most Image-CoT methods focus on text-to-image (T2I) generation. Unlike T2I generation, image editing is goal-directed: the solution space is constrained by the source image and instruction. This mismatch causes three challenges when applying Image-CoT to editing: inefficient resource allocation with fixed sampling budgets, unreliable early-stage verification using general MLLM scores, and redundant edited results from large-scale sampling. To address this, we propose ADaptive Edit-CoT (ADE-CoT), an on-demand test-time scaling framework to enhance editing efficiency and performance. It incorporates three key strategies: (1) a difficulty-aware resource allocation that assigns dynamic budgets based on estimated edit difficulty; (2) edit-specific verification in early pruning that uses region localization and caption consistency to select promising candidates; and (3) depth-first opportunistic stopping, guided by an instance-specific verifier, that terminates when intent-aligned results are found. Extensive experiments on three SOTA editing models (Step1X-Edit, BAGEL, FLUX.1 Kontext) across three benchmarks show that ADE-CoT achieves superior performance-efficiency trade-offs. With comparable sampling budgets, ADE-CoT obtains better performance with more than 2x speedup over Best-of-N.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。