Benevolent Bias in Multi-Turn Human-Agent Dialogue

2026-08-29

AI Summary

研究通过定义语气和待遇维度来检测多轮人机对话中的善意偏见，构建了BENEVDIAL语料库，并测试了现成的安全检测器和大型语言模型评判者的方法。

Abstract

Bias in human-agent interaction can manifest not only through hostile language but also as benevolent bias, whereby unequal treatment hides behind a warm, positive tone. To make it detectable, we operationalise benevolent bias along two dimensions, tone and treatment, yielding three classes: neutral support, overt bias, and benevolent bias. Building on these definitions, we construct BENEVDIAL, a class-balanced corpus of 362,880 multi-turn support dialogues spanning user and agent demographics, roles, and generators, to support controlled evaluation. We then test two detector families on it: off-the-shelf safety detectors and prompted large language model (LLM) judges. Our results reveal a detection gap: off-the-shelf detectors reliably flag overt bias yet largely miss benevolent bias, while LLM judges catch more under more explicit detection criteria but increasingly misclassify neutral support as benevolent bias, and demographic context amplifies the false alarms. These findings suggest that fair monitoring of human-agent dialogue must look beyond surface cues to whether the agent's treatment is disparate.

Problem

• benevolent bias
• human-agent dialogue
• unequal treatment

Innovation

• benevolent bias
• multi-turn dialogue
• LLM judges
• detection gap
• demographic context