Skip to content
Hot eventLive

MATE:多轮人机交互中的共情校准失效诊断

1 reports1 sources4 hr ago updated

Get the story

AI overview

2026年10月8日,arXiv(Human-Computer Interaction,一手来源)发表研究,提出 MATE(Multi-turn Assessment of calibraTed Empathy)框架,将共情视为多轮、以感知为中心的过程进行评估。两项对照研究(N=82)显示,基线回复存在对披露不敏感的校准失效,泛化式肯定与套路化语气在多轮交互中显得空洞;提示级自我批评可减弱该现象,但会引入认同偏移的校准失效,在高度人格对齐下占 Study 2 适配参与者的 62.5%。目前该事件仅有此篇报道,尚未见后续验证或争议信息。

Generated from reports · updated 3 hr ago

Timeline

Follow the coverage from different angles.

Oct 8, 2026
  1. arXiv · Human-Computer Interaction
    MATE:诊断多轮人机交互中的共情校准失效

    研究提出 MATE(Multi-turn Assessment of calibraTed Empathy)框架,把共情视为多轮、以感知为中心的过程来评估。两项对照研究(N=82)显示,基线回复存在对披露不敏感的校准失效,泛化式肯定与套路化语气在多轮中显得空洞;提示级自我批评可减弱该现象,但带来认同偏移的校准失效,在高度人格对齐下占 Study 2 适配参与者的 62.5%。

Heat trend

Current heat 9·Comparable peak 10(Oct 8)·Comparable change over 24 hours –

02.557.510Oct8Oct8Oct8Oct8

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.