Unverified50% confidenceFactExact time
McNemar检验的零假设是b和c在总体中相等，即两个模型犯错方向对称，无系统性优劣
1
Sources
50%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
McNemar检验对比机器学习模型：随机种子下的正确用法
redditr/learnmachinelearning7/11/2026
Related Claims
UnverifiedMcNemar 检验关注两个模型在每个样本上的分歧情况，只有判断结果不一致的样本（b和c单元格）具有信息量，两者都对或都错的样本被视为无关信息79% similarUnverifiedOpenAI's chart shows that at lower test-time compute levels, the model's success rate on this problem is close to zero, with occasional success only under extremely long reasoning chains.66% similarUnverified对比学习中的推远操作只要求负样本相似度低于正样本，并不要求达到-1的极端值66% similarUnverifiedThe relationship between the standard error of Monte Carlo estimation and the number of simulations N is that the standard error is proportional to 1/√N.64% similarUnverified多重比较问题指同时测试多个模型变体时，即使每个单独检验显著性水平设为0.05，整体假阳性率也会急剧膨胀64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/503209API
curl https://kongchang.com/api/v1/knowledge/claims/503209MCP
get_claim(id=503209)