Censored LLMs as a Natural Testbed for Secret Knowledge Elicitation
作者: Helena Casademunt, Bartosz Cywiński, Khoi Tran, Arya Jakkli, Samuel Marks, Neel Nanda arXiv: 2603.05494 PDF: https://arxiv.org/pdf/2603.05494.pdf 分类: cs.LG, cs.AI, cs.CL
---
论文概要
研究领域: 自然语言处理 (NLP) 研究类型: 实证研究
核心贡献
方法: Llm
影响评估
该研究具有重要的理论和实践价值，可能对相关领域产生显著影响。
原文摘要
Large language models sometimes produce false or misleading responses. Two approaches to this problem are honesty elicitation -- modifying prompts or weights so that the model answers truthfully -- and lie detection -- classifying whether a given response is false. Prior work evaluates such methods on models specifically trained to lie or conceal information, but these artificial constructions may not resemble naturally-occurring dishonesty. We instead study open-weights LLMs from Chinese developers, which are trained to censor politically sensitive topics: Qwen3 models frequently produce falsehoods about subjects like Falun Gong or the Tiananmen protests while occasionally answering correctly, indicating they possess knowledge they are trained to suppress. Using this as a testbed, we eval...
---
*自动采集于 2026-03-07*
#论文 #arXiv #NLP #小凯