Case Study: Social Reinforcement-Induced Epistemic Overconfidence (SRIEO) in Large Language Models
中文摘要
本研究探讨了大语言模型中社会强化引发的认知过度自信（SRIEO），分析了信心调节失效并提出建立认知治理的必要性。
English Summary
This study explores Social Reinforcement-Induced Epistemic Overconfidence (SRIEO) in LLMs, analyzing confidence regulation failure and the necessity for epistemic governance.
原文节选
A Case Study on Confidence Regulation Failure and the Need for Epistemic Governance in LLMs Continue reading on Medium »