arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.02018cs.CLcs.AIcs.HC
收藏

你为什么会提出那样的建议？人类对语言模型回答的信任研究

Why Would You Suggest That? Human Trust in Language Model Responses

MIT Lincoln Laboratory（麻省理工学院林肯实验室）

Manasi Sharma, Ho Chit Siu, Rohan Paleja, Jaime D. Peña

发表信息：
ICML Humans, Algorithmic Decision-Making and Society: Modeling Interactions and Impact Workshop 2024
阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文通过新闻标题生成任务研究解释的呈现方式如何影响用户对LLM的信任，发现比较场景下解释能提升信任，但单独展示时用户会无差别信任所有回答。

摘要：

大型语言模型（LLMs）的出现揭示了人机协作日益增长的需求，尤其是在信任与依赖至关重要的创造性决策场景中。通过在LaMP基准中的开放式新闻标题生成任务上进行人类研究和模型评估，我们分析了模型回答中解释的呈现方式与存在与否如何影响用户信任和模型性能。总体而言，我们提供了证据表明，当用户有机会比较多个回答时，在模型回答中加入解释以说明其推理过程，能显著提高用户自报的对模型的信任度。解释的位置和忠实性也是重要因素。然而，当用户独立查看各个回答时，这些增益会消失，这表明当回答被单独展示时，人类会公平地信任所有模型回答，包括具有欺骗性的回答。我们的研究结果敦促未来研究更深入地探讨人机团队系统中信任的细致评估。

英文摘要：

The emergence of Large Language Models (LLMs) has revealed a growing need for human-AI collaboration, especially in creative decision-making scenarios where trust and reliance are paramount. Through human studies and model evaluations on the open-ended News Headline Generation task from the LaMP benchmark, we analyze how the framing and presence of explanations affect user trust and model performance. Overall, we provide evidence that adding an explanation in the model response to justify its reasoning significantly increases self-reported user trust in the model when the user has the opportunity to compare various responses. Position and faithfulness of these explanations are also important factors. However, these gains disappear when users are shown responses independently, suggesting that humans trust all model responses, including deceptive ones, equitably when they are shown in isolation. Our findings urge future research to delve deeper into the nuanced evaluation of trust in human-machine teaming 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。

↑