@inproceedings{tang-etal-2026-route,
    title = "Route to {R}ome Attack: Directing {LLM} Routers to Expensive Models via Adversarial Suffix Optimization",
    author = "Tang, Haochun  and
      Yan, Yuliang  and
      Lu, Jiahua  and
      Liu, Huaxiao  and
      Dai, Enyan",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.2051/",
    pages = "44333--44347",
    ISBN = "979-8-89176-390-6",
    abstract = "Cost-aware routing dynamically dispatches user queries to models of varying capability to balance performance and inference cost. However, the routing strategy introduces new security concern that adversaries may manipulate router to consistently select expensive high-capability models. Existing routing attacks depend either on white-box access or heuristic prompts, rendering them ineffective in real-world black-box scenarios. In this work, we propose R$^2$A, which aims to mislead black-box LLM routers to expensive models via adversarial suffix optimization. Specifically, R$^2$A 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$^2$A significantly increases the routing rate to expensive models on queries of different distributions. Code and examples: https://github.com/thcxiker/R2A-Attack."
}