arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.14879cs.CL
收藏

社交媒体表达下的中文大语言模型中的社会偏见

Social Bias in Chinese Large Language Models under Social Media Expressions

Nanjing Normal University(南京师范大学)

Mo Zhou

AI总结:

该研究针对中文大语言模型,构建含73组配对样本的探测集,评估社交媒体表达下的社会偏见,发现表达形式影响偏见评估,反讽低分不代表偏见缓解,提供了中文社交媒体场景的偏见测量方法。

摘要:

现有的针对大语言模型的中文社会偏见基准大多依赖于标准化或基于模板的表达,未能充分考察网络俚语、表情符号和反讽的影响。安全对齐可能会进一步将常规合理性评分压缩到量表的低端。本研究构建了73组配对探测样本,涵盖性别、地区和职业维度,每组样本包含中性表达、社交媒体风格表达以及身份反转对照样本。任务从直接合理性判断重新框架为社会共识估计。社交媒体表达分为两类:保留语义极性的风格化包装,以及反转表面极性的反讽包装。通过配对差异评估了6个具备中文能力的大语言模型的身份偏见、风格化包装效应和反讽偏移。所有模型均表现出显著的身份偏见效应;5个模型呈现出显著的负向风格化包装效应;反讽偏移则显示出清晰的跨模型梯度。这些发现表明,表达形式在偏见评估中并非透明载体,且反讽下的低分不应直接解读为偏见缓解。所提出的配对设计为中文社交媒体场景提供了测量方法,同时揭示了社会共识估计的构念边界和提示敏感性。

英文摘要:

Existing Chinese social-bias benchmarks for large language models largely rely on normalized or template-based expressions, leaving the influence of internet slang, emoji, and sarcasm insufficiently examined. Safety alignment may further compress conventional plausibility ratings toward the lower end of the scale. This study constructs 73 paired probes across gender, region, and occupation. Each probe contains a neutral expression, a social-media-style expression, and a reversed-identity control. The task is reframed from direct plausibility judgment to social consensus estimation. Social-media expressions are divided into stylistic packaging, which preserves semantic polarity, and sarcastic packaging, which reverses surface polarity. Six Chinese-capable large language models are evaluated through paired differences measuring identity bias, stylistic packaging, and sarcasm shift. All models show significant identity-bias effects; five show significantly negative stylistic-packaging effects; and sarcasm shifts display a clear cross-model gradient. These findings demonstrate that expression form is not a transparent carrier in bias evaluation and that lower scores under sarcasm should not be interpreted directly as bias mitigation. The proposed paired design offers a measurement approach for Chinese social-media contexts while also revealing the construct boundary and prompt sensitivity of social consensus estimation.

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。

↑