LLM能否可靠识别自身被对抗性前缀攻击？研究检验其内省能力在安全场景中的表现。
arXiv:2606.23671v1 Announce Type: new Abstract: Prior work shows that large language models (LLMs) exhibit introspective capability on benign tasks. W…
LLM心理测量评估新视角：揭示自我报告预测行为的条件与原因，为模型行为理解提供理论支撑。
arXiv:2606.12730v1 Announce Type: new Abstract: Anticipating LLM behavioral tendencies from low-cost psychometric probes is critical for safe deployme…
LLM自述心理特质与实际行为大相径庭，25个模型验证存在“言行不一”的自我报告-行为鸿沟。
arXiv:2606.09843v1 Announce Type: cross Abstract: Large language models (LLMs) produce stable self-reports on personality inventories, but these self-…
用生成式投射测试替代自我报告，让LLM心理测量更可靠，摆脱语料污染与方向性偏差
arXiv:2606.00860v1 Announce Type: cross Abstract: Self-report questionnaires remain the prevailing tool for probing the psychological states of person…