Route to Rome Attack: Directing LLM Routers to Expensive Models via Adversarial Suffix Optimization
arXiv SecurityArchived Apr 17, 2026✓ Full text saved
arXiv:2604.15022v1 Announce Type: new Abstract: Cost-aware routing dynamically dispatches user queries to models of varying capability to balance performance and inference cost. However, the routing strategy introduces a new security concern that adversaries may manipulate the router to consistently select expensive high-capability models. Existing routing attacks depend on either white-box access or heuristic prompts, rendering them ineffective in real-world black-box scenarios. In this work, w
Full text archived locally
✦ AI Summary· Claude Sonnet
Computer Science > Cryptography and Security
[Submitted on 16 Apr 2026]
Route to Rome Attack: Directing LLM Routers to Expensive Models via Adversarial Suffix Optimization
Haochun Tang, Yuliang Yan, Jiahua Lu, Huaxiao Liu, Enyan Dai
Cost-aware routing dynamically dispatches user queries to models of varying capability to balance performance and inference cost. However, the routing strategy introduces a new security concern that adversaries may manipulate the router to consistently select expensive high-capability models. Existing routing attacks depend on either white-box access or heuristic prompts, rendering them ineffective in real-world black-box scenarios. In this work, we propose R^2A, which aims to mislead black-box LLM routers to expensive models via adversarial suffix optimization. Specifically, R^2A deploys a hybrid ensemble surrogate router to mimic the black-box router. A suffix optimization algorithm is further adapted for the ensemble-based surrogate. Extensive experiments on multiple open-source and commercial routing systems demonstrate that {R^2A} significantly increases the routing rate to expensive models on queries of different distributions. Code and examples: this https URL.
Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
Cite as: arXiv:2604.15022 [cs.CR]
(or arXiv:2604.15022v1 [cs.CR] for this version)
https://doi.org/10.48550/arXiv.2604.15022
Focus to learn more
Journal reference: ACL 2026 Main Conference
Submission history
From: Yuliang Yan [view email]
[v1] Thu, 16 Apr 2026 13:51:48 UTC (523 KB)
Access Paper:
HTML (experimental)
view license
Current browse context:
cs.CR
< prev | next >
new | recent | 2026-04
Change to browse by:
cs
cs.AI
cs.CL
cs.LG
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Export BibTeX Citation
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Demos
Related Papers
About arXivLabs
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)