无关提示竟能操控大模型行为，揭示新型敏感性机制。
Spurious Prompts: Can Irrelevant Prompts Steer Large Language Models?
- 发现语义无关提示可有效引导模型决策
- 在多个基准上提升性能，媲美甚至超越标准提示
- 适用于研究模型安全与可控性，适合对齐方向读者
大型语言模型对提示高度敏感，但现有研究多聚焦于任务相关的指令、示范或推理线索。本文探讨另一种提示敏感性：语义上与任务无关的提示是否仍能引导模型行为。我们称之为伪提示（spurious prompts），并证明其出人意料的有效性。提出一种简单的黑盒搜索方法用于发现此类提示。在涵盖0.8B至27B参数、三种模型家族的推理与问答基准上，伪提示显著提升性能，常达到或超过标准提示基线及任务感知提示优化效果。进一步发现，它们可诱导模型产生非预期行为，如反复选择首个选项、生成错误答案，或输出偶数、质数或小数值，而无需明确指令。结果揭示了一种新型提示敏感性：大模型可被与任务无关的提示系统性地操控。代码已开源。
原文摘要 · Abstract (English)
Large language models are highly sensitive to prompts, but this sensitivity is usually studied through task-relevant instructions, demonstrations, or reasoning cues. In this paper, we study a different form of prompt sensitivity: whether prompts that are semantically unrelated to the task can nevertheless steer model behavior. We call them spurious prompts and show their surprising efficacy. We also propose a simple black-box search procedure for discovering them. Across reasoning and question-answering benchmarks, using models ranging from 0.8B to 27B parameters and spanning three model families, we show that spurious prompts can improve performance, often matching or outperforming standard prompting baselines and task-aware prompt optimization. We further show that they can steer models toward unintended behaviors, such as repeatedly selecting the first answer option, producing incorrect answers, returning an even, prime or small number without explicitly instructing the model to do so. These findings reveal a new kind of prompt sensitivity: LLMs can be systematically steered by prompts that are unrelated to the task they are asked to solve. Our code is available at https://github.com/Batorskq/spurious
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。