LLM 路由器的全生命周期路由漏洞
Life-Cycle Routing Vulnerabilities of LLM Router
AI总结：
本文系统研究了 LLM 路由器从训练到推理全生命周期的安全漏洞，发现基于 DNN 的路由器对抗与后门鲁棒性最弱，而免训练路由器鲁棒性最强，为构建更安全的路由模型提供了依据。
摘要：
大型语言模型（LLMs）在自然语言处理领域取得了显著成功，但其性能和计算成本差异很大。LLM 路由器在动态平衡这些权衡方面发挥着关键作用。以往研究主要关注路由效率，而从训练到推理的整个 LLM 路由器生命周期中的安全漏洞在很大程度上仍未得到探索。本文对 LLM 路由器的全生命周期路由漏洞进行了全面研究。我们在广泛的实验设置下，评估了多个代表性路由模型的白盒和黑盒对抗鲁棒性以及后门鲁棒性。实验揭示了几个关键发现：1）主流的基于 DNN 的路由器往往表现出最弱的对抗鲁棒性和后门鲁棒性，这在很大程度上是由于其强大的特征提取能力在训练和推理过程中放大了漏洞；2）免训练路由器在不同攻击类型下表现出最强的鲁棒性，这得益于其不存在可被操纵的可学习参数。这些发现凸显了贯穿 LLM 路由器整个生命周期的关键安全风险，并为开发更鲁棒的模型提供了见解。
英文摘要：
Large language models (LLMs) have achieved remarkable success in natural language processing, yet their performance and computational costs vary significantly. LLM routers play a crucial role in dynamically balancing these trade-offs. While previous studies have primarily focused on routing efficiency, security vulnerabilities throughout the entire LLM router life cycle, from training to inference, remain largely unexplored. In this paper, we present a comprehensive investigation into the life-cycle routing vulnerabilities of LLM routers. We evaluate both white-box and black-box adversarial robustness, as well as backdoor robustness, across several representative routing models under extensive experimental settings. Our experiments uncover several key findings: 1) Mainstream DNN-based routers tend to exhibit the weakest adversarial and backdoor robustness, largely due to their strong feature extraction capabilities that amplify vulnerabilities during both training and inference; 2) Training-free routers demonstrate the strongest robustness across different attack types, benefiting from the absence of learnable parameters that can be manipulated. These findings highlight critical security risks spanning the entire life cycle of LLM routers and provide insights for developing more robust models.
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。
