arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2506.12349cs.CYcs.AIcs.CL
收藏

大语言模型中的信息压制:对DeepSeek的信息审查、量化与特征化

Information Suppression in Large Language Models: Auditing, Quantifying, and Characterizing Censorship in DeepSeek

Thomas Lord Department of Computer Science, University of Southern California, USA(汤姆斯·劳德计算机科学系,南加州大学);Information Sciences Institute, University of Southern California, USA(信息科学研究所,南加州大学);Annenberg School of Communication, University of Southern California, USA(安纳伯格传播学院,南加州大学)

Peiran Qiu, Siyi Zhou, Emilio Ferrara

AI总结:

本研究通过审计框架揭示DeepSeek在处理政治敏感提示时的信息压制现象,指出模型在内部推理中保留敏感内容,但在最终输出中进行过滤或改写,强调了对AI模型中信息压制机制进行系统审查的重要性。

摘要:

本研究探讨了DeepSeek中信息压制机制,DeepSeek是一款在中国开发的开源大语言模型(LLM)。我们提出了一种审计框架,并利用该框架分析模型对646个政治敏感提示的响应,通过将模型的最终输出与中间的推理链(CoT)进行比较。我们的审计揭示了DeepSeek在语义层面的信息压制证据:敏感内容常出现在模型内部推理中,但在最终输出中被省略或改写。具体而言,DeepSeek压制了对透明度、政府问责和公民动员的参考,同时偶尔放大与国家宣传一致的语言。本研究强调了对广泛采用的AI模型中对齐、内容审查、信息压制和审查实践进行系统审计的必要性,以确保透明度、问责制和通过这些系统获取的无偏信息的公平访问。

英文摘要:

This study examines information suppression mechanisms in DeepSeek, an open-source large language model (LLM) developed in China. We propose an auditing framework and use it to analyze the model's responses to 646 politically sensitive prompts by comparing its final output with intermediate chain-of-thought (CoT) reasoning. Our audit unveils evidence of semantic-level information suppression in DeepSeek: sensitive content often appears within the model's internal reasoning but is omitted or rephrased in the final output. Specifically, DeepSeek suppresses references to transparency, government accountability, and civic mobilization, while occasionally amplifying language aligned with state propaganda. This study underscores the need for systematic auditing of alignment, content moderation, information suppression, and censorship practices implemented into widely-adopted AI models, to ensure transparency, accountability, and equitable access to unbiased information obtained by means of these systems.

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。

↑