Unverified50% confidenceFactExact time
The study introduces an information processing gap metric to measure the deviation between a model's actual belief updates and ideal Bayesian updates.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe core finding of the study is that there is a systematic deviation between large models' belief updates and Bayesian rules.79% similarUnverifiedThe bias-variance tradeoff is the core framework for describing a model's generalization capability: bias measures the systematic deviation of the model's predictions from the true values (underfitting), while variance measures the fluctuation of the model's predictions across different training sets (overfitting).69% similarUnverifiedWhen prepending is applied to models such as Kimi and GLM, the distribution of reasoning length undergoes a statistically significant shift and aligns toward a certain source model.67% similarUnverified部分研究者认为大模型涌现能力是真实的相变现象，另一些人认为这只是评估指标选择造成的统计假象67% similarUnverified要求模型列出反直觉假设、从批评者角度反驳主流观点或连接两个看似无关领域，可以激活模型更边缘的知识表示67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/897398API
curl https://kongchang.com/api/v1/knowledge/claims/897398MCP
get_claim(id=897398)