Unverified50% confidenceFactExact time

The study evaluated AI metanorm reasoning along two dimensions—emotional appraisal and behavioral response—where behavioral response included two tasks: predicting the violator's self-regulation and bystander other-regulation.

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

9/11/2026

First Seen

Valid until: 12/10/2026

Sources

AI's Social Cognition Deficit: Why Models Consistently Over-Punish

Related Entities

NormReactproduct 元规范推理concept

Related Claims

Unverified该会议总结Agent Skill内置了情感分析模块，能判断会议整体的情感倾向为积极、消极或中性79% similar Unverified情感分析、用户意图识别和转化节点检测属于NLP领域不同子任务，单一模型难以在所有子任务上都表现最优73% similar UnverifiedAI直接描述情绪在统计上是更"安全"、更高频的表达选择，符合模型对"好写作"的概率预测72% similar Unverified方面级情感分析（ABSA）能够从一条评论中同时提取多个评价维度和情感极性71% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/892947

API

curl https://kongchang.com/api/v1/knowledge/claims/892947

MCP

get_claim(id=892947)