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 2402.01789cs.CYcs.AIcs.CL
收藏

LLM的政治偏好

The Political Preferences of LLMs

Otago Polytechnic(奥塔哥理工学院)

David Rozado

AI总结:

本文通过对24个对话式LLM进行11项政治倾向测试,发现主流LLM普遍表现出中左翼偏好,并证明可通过监督微调(SFT)将LLM引导至政治光谱特定位置,揭示了其社会影响。

摘要:

我在此报告了对大型语言模型(LLM)中嵌入的政治偏好的综合分析。具体而言,我对24个最先进的对话式LLM(包括闭源和开源)实施了11项旨在识别测试者政治偏好的政治倾向测试。当使用带有政治内涵的问题或陈述进行探测时,主流对话式LLM倾向于生成被政治测试工具诊断为表现出中左翼偏好的回答。对于五个基础模型(即构建LLM的基础模型),情况似乎并非如此,因为优化用于与人类对话的LLM是建立在这些基础模型之上的。基础模型在连贯回答测试问题方面的较弱表现使得这部分结果缺乏说服力。我证明了通过仅使用适量政治对齐数据的监督微调(SFT),就可以将LLM引导至政治光谱中的特定位置,这表明SFT具有在LLM中嵌入政治倾向的潜力。随着LLM开始部分取代搜索引擎、Wikipedia等传统信息源,LLM中嵌入的政治偏见的社会影响是巨大的。

英文摘要:

I report here a comprehensive analysis about the political preferences embedded in Large Language Models (LLMs). Namely, I administer 11 political orientation tests, designed to identify the political preferences of the test taker, to 24 state-of-the-art conversational LLMs, both closed and open source. When probed with questions/statements with political connotations, most conversational LLMs tend to generate responses that are diagnosed by most political test instruments as manifesting preferences for left-of-center viewpoints. This does not appear to be the case for five additional base (i.e. foundation) models upon which LLMs optimized for conversation with humans are built. However, the weak performance of the base models at coherently answering the tests' questions makes this subset of results inconclusive. Finally, I demonstrate that LLMs can be steered towards specific locations in the political spectrum through Supervised Fine-Tuning (SFT) with only modest amounts of politically aligned data, suggesting SFT's potential to embed political orientation in LLMs. With LLMs beginning to partially displace traditional information sources like search engines and Wikipedia, the societal implications of political biases embedded in LLMs are substantial.

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。

↑