cs.LGOct 30, 2026

RetryRoute: Retry-Aware Model Routing for Agentic Image Editing Services

Authors: Hanshuai Cui, Zhiqing Tang, Zhi Yao, Jun Luo, Weijia Jia

Organizations: School of Artificial Intelligence, Beijing Normal University, Beijing 100875, China · Institute of Artificial Intelligence and Future Networks, Beijing Normal University, Zhuhai 519087, China · College of Computing and Data Science, Nanyang Technological University, Singapore

Abstract

Web agents increasingly edit images by calling open editing models in a critique-and-retry loop, and these models differ in cost by more than an order of magnitude. After every rejected edit, a service must decide whether to retry the same model, escalate to another or give up. Existing routers pick one model per request, and cascades escalate in a fixed order. However, both treat a retry as an independent draw. We identify sticky failures, meaning that a retry after a failure succeeds only 17-53% as often as a first try across six open models, three benchmarks and two judges. Based on this finding, we propose RetryRoute, a retry-aware router that estimates each model's chance of success given the request's failure history. It plans retries, escalations and abandonment over failure states and adds live queue waits to the cost, so the same plan becomes load-aware dispatch. At the main budget, RetryRoute's success under the in-loop judge exceeds the strongest of eleven prior baselines by 2.8 points on average over eight groups. On a GPU serving testbed it raises the share of edits delivered within the service-level objective by 5.0-6.3 points over the best prior baseline at every load.

Figures & tables

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. EditJev: System One Routing for Image Editing Models

    Oct 23, 2026Zhile Guan, Xu Jiang, Yule Duan +4

  2. EditAgent: Rethinking Agentic Image Editing

    Oct 29, 2026Haiyang Xu, Tianyu Wang, Jialu Li +5Image EditingTool-Using Vision-Language Agents

  3. Disagreement-Guided Weak-to-Strong Routing

    Oct 15, 2026Sima Noorani, Shayan Kiyani, George Pappas +1LLM Routing