17:46
2026-09-11
research.nvidia.com
ai-safety
ReasoningBomb: A Stealthy Denial-of-Service Attack by Inducing Pathologically Long Reasoning in Large Reasoning Models
Researchers presented ReasoningBomb, a reinforcement-learning-based prompt-induced inference-time denial-of-service (PI-DoS) framework that drives victim large reasoning models into pathologically lon…