← Back to all news
arXiv AI By Tianyi Wang, Huawei Fan, Yuanchao Shu, Peng Cheng, Cong Wang

Rethinking Latency Denial-of-Service: Attacking the LLM Serving Framework, Not the Model

Read the original on arXiv AI →

The paper "Rethinking Latency Denial-of-Service: Attacking the LLM Serving Framework, Not the Model" argues that traditional algorithmic latency attacks are largely ineffective against modern LLM serving systems due to system-level optimizations like continuous batching. It introduces a new Fill and Squeeze attack that targets the scheduler’s state transition, exhausting the global KV cache to cause Head-of-Line blocking and forcing repetitive preemption. Experiments on vLLM show the attack can cause up to 75–742× TTFT degradation and 1.5–4× slowdown on Time Per Output Token, achieving these effects with 30–40% lower attack cost.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv AI
Sep 25

Persistent Billable State: Denial-of-Wallet Attacks and Defenses in Tool-Calling LLM Agents

The paper investigates how multi‑step tool‑calling large language model agents can unintentionally create persistent billable state, allowing malicious or compromised tools to generate repeated charges without user credentials. It formalizes the persistent billable‑state boundary, identifies six denial‑of‑wallet attack vectors, and evaluates them with the DOW‑BENCH harness across six model families. The study shows significant cost amplification, demonstrates effective mitigation via deterministic history transformation and host‑side invariants, and highlights the scarcity of existing safeguards in real‑world repositories.

By Jinqian Zhang (Institute of Information Engineering, Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences), Haojun Xia (Institute of Information Engineering, Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences), Shujiang Wu (Beihang University), Jingkun Yue (State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing, China), Xia Zhang (Institute of Information Engineering, Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences), Zhangpei Cheng (Institute of Information Engineering, Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences), Bibo Tu (Institute of Information Engineering, Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences)
llmsagents
More like this →