Unverified50% confidenceFactExact time
The statistical intuition of pooling: merging the generation counts of all questions to estimate shared weights is equivalent to using more data to estimate fewer parameters to reduce variance, but when the relative merits of models differ greatly across questions, forced sharing introduces bias.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedPooling is superior to per-problem tuning: when there is a common shape to borrow across problems, sharing information can significantly reduce overall calibration error, while tuning each problem independently incurs a clear cost when labels are few.78% similarUnverifiedMechanism two: in the intermediate regime, pooling the generation counts of different questions is more effective than estimating each answer distribution separately, and only logarithmic gains can be obtained from additional questions.74% similarUnverifiedThe more independent requirements included in a single prompt, the more the probability of the model omitting or incorrectly implementing certain requirements grows non-linearly; the MVP strategy improves reliability by narrowing the model's decision space.73% similarUnverified多模型集成能平滑单一模型的系统性偏差，多模型高度一致时结论可信度更高，显著分歧时提示不确定性需人工核查73% similarUnverifiedUsing the median rather than the mean is to avoid disturbance of the metric by long-tail samples and to better reflect the handling experience of typical issues.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1046171API
curl https://kongchang.com/api/v1/knowledge/claims/1046171MCP
get_claim(id=1046171)