知识盲区：一种用于审计LLM辅助分析中先验污染的推理时协议
Epistemic Blinding: An Inference-Time Protocol for Auditing Prior Contamination in LLM-Assisted Analysis
AI总结：
本文提出了一种在代理系统中用于审计LLM辅助分析中先验污染的推理时协议，通过替换实体标识符以匿名代码并比较输出与未盲对照组，以恢复审计性。
摘要：
本文提出知识盲区，一种在使用大型语言模型跨多个生物数据集进行推理的代理系统中用于审计先验污染的推理时协议。在开发过程中发现，LLM输出会无声地将数据驱动的推断与记忆中的先验知识混合，且这种混合是不可见的：无法从单个输出确定其中有多少来自页面上的数据，有多少来自模型的训练记忆。知识盲区是一种简单的推理时协议，它在提示时用匿名代码替换实体标识符，然后将输出与未盲对照组进行比较。该协议不使LLM推理确定性，但恢复了审计性的一个关键轴：测量输出中来自提供的数据与模型参数知识的比例。完整的目标识别系统被描述——包括LLM引导的评分函数进化优化和盲目的代理推理用于目标合理性——并证明两个阶段均无法访问实体身份。在四种癌症类型的肿瘤药物靶点优先排序中，盲化使前20个预测中的16%发生变化，同时保持验证靶点的恢复相同。污染问题被证明不仅限于生物学：在S&P 500股票筛选中，品牌识别偏见重塑了五个随机种子中的30-40%的前20名排名。为降低采用门槛，该协议作为开源工具和Claude Code技能发布，使代理工作流中的一条命令即可实现知识盲化。主张并非盲化分析会产生更好结果，而是没有盲化时，无法确定代理在多大程度上遵循研究人员设计的分析过程。
英文摘要：
This paper presents epistemic blinding in the context of an agentic system that uses large language models to reason across multiple biological datasets for drug target prioritization. During development, it became apparent that LLM outputs silently blend data-driven inference with memorized priors about named entities - and the blend is invisible: there is no way to determine, from a single output, how much came from the data on the page and how much came from the model's training memory. Epistemic blinding is a simple inference-time protocol that replaces entity identifiers with anonymous codes before prompting, then compares outputs against an unblinded control. The protocol does not make LLM reasoning deterministic, but it restores one critical axis of auditability: measuring how much of an output came from the supplied data versus the model's parametric knowledge. The complete target identification system is described - including LLM-guided evolutionary optimization of scoring functions and blinded agentic reasoning for target rationalization - with demonstration that both stages operate without access to entity identity. In oncology drug target prioritization across four cancer types, blinding changes 16% of top-20 predictions while preserving identical recovery of validated targets. The contamination problem is shown to generalize beyond biology: in S&P 500 equity screening, brand-recognition bias reshapes 30-40% of top-20 rankings across five random seeds. To lower the barrier to adoption, the protocol is released as an open-source tool and as a Claude Code skill that enables one-command epistemic blinding within agentic workflows. The claim is not that blinded analysis produces better results, but that without blinding, there is no way to know to what degree the agent is adhering to the analytical process the researcher designed.
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。
