Inducing Overthink: Hierarchical Genetic Algorithm-based DoS Attack on Black-Box Large Language Reasoning Models
Researchers have developed a new method to exploit a vulnerability in large language reasoning models (LRMs) that causes them to "overthink." This technique uses a hierarchical genetic algorithm to generate inputs that lead to excessively long and redundant reasoning traces, increasing latency and resource consumption. The attack demonstrated significant increases in output length, up to 26.1x on the MATH benchmark, and showed effectiveness against various state-of-the-art models, highlighting a need for improved defenses against such denial-of-service attacks. AI
IMPACT This research reveals a new vulnerability in LLM reasoning, potentially impacting the reliability and availability of AI systems that depend on them.