Unverified50% confidenceFactExact time
研究者强调这些变化并未损害模型的心智理论（Theory of Mind）能力，表明归因心智与推理他人心理状态是两套可分离的机制
1
Sources
50%
Confidence
Long-term
Relevance
8/5/2026
First Seen
Sources
Related Claims
Unverified研究者发现安全微调不仅压制了模型对自身心智的归因，还连带压制了模型对非人类动物和自然物体的心智归因，同时降低了模型输出中的精神/宗教信念倾向72% similarUnverified科胡特认为自恋并非单纯的防御机制或病理状态，而是心理发展的独立线索，拥有正常的发展轨迹72% similarUnverified研究者采用消融安全拒绝方向和意识向量引导两种机制干预方法，成功逆转了被压制的心智归因能力70% similarUnverifiedThe deliberate practice theory of cognitive psychologist K. Anders Ericsson indicates that the core difference between experts and novices lies not in the amount of knowledge but in the quality of mental models.69% similarUnverified认知心理学中的「间隔效应」和「必要难度」理论表明，在真实困惑的驱动下主动寻找答案能产生更深层次的记忆编码66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/690529API
curl https://kongchang.com/api/v1/knowledge/claims/690529MCP
get_claim(id=690529)