Unverified50% confidenceTradeoffExact time
Softmax tends to produce sharp, winner-takes-all distributions, whereas the distribution shape of squared normalization may be more spread out or more concentrated.
1
Sources
50%
Confidence
Long-term
Relevance
10/10/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedSoftmax函数具有平移不变性，任意放大所有logit的尺度不改变预测类别，却会将输出概率压向极端值，是其天然的过度自信倾向71% similarUnverified计算 softmax 函数时指数运算可能导致浮点数溢出，标准做法是先减去输入向量的最大值再取指数71% similarUnverifiedWhen a model encounters out-of-distribution inputs, prediction entropy increases, the softmax output tends toward a uniform distribution, and the range of candidate tokens permitted by nucleus sampling expands, giving long-tail anomalous tokens a chance of being sampled.69% similarUnverifiedSoftmax输出的概率值并不等同于真实的贝叶斯后验概率，而是被过度优化以拉开类别间距，导致预测概率系统性偏高69% similarUnverifiedSoftmax函数将神经网络logits压缩为概率分布，但其输出值在大规模模型中系统性偏离真实贝叶斯后验概率68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/995032API
curl https://kongchang.com/api/v1/knowledge/claims/995032MCP
get_claim(id=995032)