Sheta, O., Deuja, R., Masoudi, H., & Fang, M. (2026). Does the Unsafe Gradient Survive a Conversation? On the Fragility of Gradient-Based Jailbreak Detection in Multi-Turn Dialogue. https://omanscience.com/ar/articles/does-the-unsafe-gradient-survive-a-conversation-on-the-fragility-of-gradient-based-jailbreak-detection-in-multi-turn-dialogue