← 全部论文

Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems

整体是否超过部分之和?AI 解释对互补团队表现的影响

英文原题: Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance

2021-05-06541 次引用DOI 10.1145/3411764.3445717

Gagan Bansal · Tongshuang Wu · Joyce Zhou · Raymond Fok · Besmira Nushi · Ece Kamar · Marco Tulio Ribeiro · Daniel S. Weld

核心结论

基于三个数据集的混合方法研究发现,AI 辅助能带来表现提升,但解释并未进一步提高人机互补表现;解释反而会增加人们接受 AI 建议的概率,无论建议正确与否。

AI 生成的 HIE 评估

研究设计

mixed method user study

title and abstract

研究发现

Mixed-method user studies across three datasets tested whether AI explanations help human–AI teams outperform either humans or AI alone. The studies observed complementary gains from AI assistance, but explanations did not increase those gains. Explanations instead made people more likely to accept AI recommendations whether they were correct or not, highlighting the difference between explainability and appropriately calibrated reliance.

原论文记录

证据强度

证据优势

  • 混合方法用户研究
  • 覆盖三个数据集
  • 区分建议接受度与建议正确性

局限性

  • 任务领域有限
  • 解释方式与界面未必代表新一代生成式 AI

这项研究不能证明什么

本记录不能证明结论适用于论文所述人群、任务、时间范围与方法之外的情境,也不应被解读为普遍因果结论。

HIE 解读

Tests whether explanations produce genuinely complementary team performance rather than indiscriminate reliance on AI advice.

AI 生成的 HIE 评估 · AI 评估置信度 88

相关主题

  • Human-AI Interaction
  • Artificial Intelligence
  • Social Intelligence

关联论断

原始来源

DOI 10.1145/3411764.3445717 ↗