{"ID":2846130,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.02376","arxiv_id":"2511.02376","title":"AutoAdv: Automated Adversarial Prompting for Multi-Turn Jailbreaking of Large Language Models","abstract":"Large Language Models (LLMs) remain vulnerable to jailbreaking attacks where adversarial prompts elicit harmful outputs. Yet most evaluations focus on single-turn interactions while real-world attacks unfold through adaptive multi-turn conversations. We present AutoAdv, a training-free framework for automated multi-turn jailbreaking that achieves an attack success rate of up to 95% on Llama-3.1-8B within six turns, a 24% improvement over single-turn baselines. AutoAdv uniquely combines three adaptive mechanisms: a pattern manager that learns from successful attacks to enhance future prompts, a temperature manager that dynamically adjusts sampling parameters based on failure modes, and a two-phase rewriting strategy that disguises harmful requests and then iteratively refines them. Extensive evaluation across commercial and open-source models (Llama-3.1-8B, GPT-4o mini, Qwen3-235B, Mistral-7B) reveals persistent vulnerabilities in current safety mechanisms, with multi-turn attacks consistently outperforming single-turn approaches. These findings demonstrate that alignment strategies optimized for single-turn interactions fail to maintain robustness across extended conversations, highlighting an urgent need for multi-turn-aware defenses.","short_abstract":"Large Language Models (LLMs) remain vulnerable to jailbreaking attacks where adversarial prompts elicit harmful outputs. Yet most evaluations focus on single-turn interactions while real-world attacks unfold through adaptive multi-turn conversations. We present AutoAdv, a training-free framework for automated multi-tur...","url_abs":"https://arxiv.org/abs/2511.02376","url_pdf":"https://arxiv.org/pdf/2511.02376v3","authors":"[\"Aashray Reddy\",\"Andrew Zagula\",\"Nicholas Saban\"]","published":"2025-11-04T08:56:28Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.AI\",\"cs.CR\",\"cs.LG\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}