arXiv 2310.125582024-04-01cs.CLcs.HC

大型语言模型帮助人类验证真实性——除非它们以令人信服的方式出错

Large Language Models Help Humans Verify Truthfulness -- Except When They Are Convincingly Wrong

Stanford University（斯坦福大学）；University of Maryland（马里兰大学）；Carnegie Mellon University（卡内基梅隆大学）；NYU Shanghai（上海纽约大学）；New York University（纽约大学）；Microsoft Research（微软研究院）

Chenglei Si, Navita Goyal, Sherry Tongshuang Wu, Chen Zhao, Shi Feng, Hal Daumé, Jordan Boyd-Graber

备注：: NAACL 2024

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文通过80名众包实验发现LLM解释虽能提升事实核查效率但易致用户过度依赖，对比性解释可缓解该问题却未能超越搜索引擎，表明LLM解释在高风险场景中不可靠。

摘要：

大型语言模型（LLMs）越来越多地被用于访问网络信息，因此其真实性和事实性备受关注。为了帮助用户对其获取的信息做出正确决策，LLMs不仅应提供信息，还应帮助用户进行事实核查。我们与80名众包工作者进行的实验比较了语言模型与搜索引擎（信息检索系统）在促进事实核查方面的表现。我们提示LLMs验证给定声明并提供相应解释。阅读LLM解释的用户比使用搜索引擎的用户效率显著更高，同时达到相似的准确率。然而，当解释错误时，用户会过度依赖LLMs。为了减少对LLMs的过度依赖，我们要求LLMs提供对比性信息——同时解释声明为何为真以及为何为假，然后将解释的两面都呈现给用户。这种对比性解释缓解了用户对LLMs的过度依赖，但无法显著优于搜索引擎。此外，同时展示搜索引擎结果和LLM解释相比单独使用搜索引擎并没有提供互补性收益。综上所述，我们的研究强调，LLMs的自然语言解释可能不是阅读检索到的段落的可靠替代品，尤其是在高风险场景中，过度依赖错误的人工智能解释可能导致严重后果。

英文摘要：

Large Language Models (LLMs) are increasingly used for accessing information on the web. Their truthfulness and factuality are thus of great interest. To help users make the right decisions about the information they get, LLMs should not only provide information but also help users fact-check it. Our experiments with 80 crowdworkers compare language models with search engines (information retrieval systems) at facilitating fact-checking. We prompt LLMs to validate a given claim and provide corresponding explanations. Users reading LLM explanations are significantly more efficient than those using search engines while achieving similar accuracy. However, they over-rely on the LLMs when the explanation is wrong. To reduce over-reliance on LLMs, we ask LLMs to provide contrastive information - explain both why the claim is true and false, and then we present both sides of the explanation to users. This contrastive explanation mitigates users' over-reliance on LLMs, but cannot significantly outperform search engines. Further, showing both search engine results and LLM explanations offers no complementary benefits compared to search engines alone. Taken together, our study highlights that natural language explanations by LLMs may not be a reliable replacement for reading the retrieved passages, especially in high-stakes settings where over-relying on wrong AI explanations could lead to critical consequences.

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。