永远提问:对话LLM中轮流放大背后的通用激活
Asking Forever: Universal Activations Behind Turn Amplification in Conversational LLMs
AI总结:
研究提出对话LLM中一种新的失败模式'轮流放大',通过机制视角识别通用激活子空间,展示攻击者利用澄清寻求行为延长交互,且现有防御措施效果有限。
摘要:
多轮交互长度是对话LLM运营成本的主要因素。在本工作中,我们提出了一种新的失败模式:轮流放大,即模型会持续延长多轮交互而不完成底层任务。我们显示,攻击者可以系统地利用澄清寻求行为——在多轮对话设置中普遍鼓励的行为——来大规模延长交互。超越提示层面的行为,我们从机制角度出发,识别出一个与澄清寻求响应相关的查询无关的通用激活子空间。不同于以往依赖每轮提示优化的成本放大攻击,我们的攻击源于对话动态,并在提示和任务之间持续存在。我们展示这种机制提供了一条可扩展的路径来诱导轮流放大:通过供应链攻击通过微调和运行时攻击通过低层参数破坏,一致地将模型转向抽象、澄清寻求行为。在多个指令微调的LLM和基准测试中,我们的攻击显著增加轮次计数,同时保持合规。我们还显示现有防御措施对这种新兴失败类别的保护有限。
英文摘要:
Multi-turn interaction length is a dominant factor in the operational costs of conversational LLMs. In this work, we present a new failure mode in conversational LLMs: turn amplification, in which a model consistently prolongs multi-turn interactions without completing the underlying task. We show that an adversary can systematically exploit clarification-seeking behavior$-$commonly encouraged in multi-turn conversation settings$-$to scalably prolong interactions. Moving beyond prompt-level behaviors, we take a mechanistic perspective and identify a query-independent, universal activation subspace associated with clarification-seeking responses. Unlike prior cost-amplification attacks that rely on per-turn prompt optimization, our attack arises from conversational dynamics and persists across prompts and tasks. We show that this mechanism provides a scalable pathway to induce turn amplification: both supply-chain attacks via fine-tuning and runtime attacks through low-level parameter corruptions consistently shift models toward abstract, clarification-seeking behavior across prompts. Across multiple instruction-tuned LLMs and benchmarks, our attack substantially increases turn count while remaining compliant. We also show that existing defenses offer limited protection against this emerging class of failures.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。