Paper Detail

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

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

arxiv Score 14.5

Published 2026-10-30 · First seen 2026-11-02

Research Track B · General AI

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.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{cui2026retryroute,
  title = {RetryRoute: Retry-Aware Model Routing for Agentic Image Editing Services},
  author = {Hanshuai Cui and Zhiqing Tang and Zhi Yao and Jun Luo and Weijia Jia},
  year = {2026},
  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 17-5},
  url = {https://arxiv.org/abs/2610.41460},
  keywords = {cs.LG},
  eprint = {2610.41460},
  archiveprefix = {arXiv},
}

Metadata

{}