Unverified50% confidenceFactExact time
Scaling Laws表明模型的交叉熵损失会随着计算量、数据量和参数量的增加而平滑下降，且这种下降关系可以用简单的幂律函数精确拟合
1
Sources
50%
Confidence
Long-term
Relevance
8/28/2026
First Seen
Sources
Related Entities
Related Claims
Unverified如果交叉熵稳步下降而Dice波动剧烈，问题几乎可以确定出在Dice损失的配置上78% similarUnverifiedScaling Laws found that the cross-entropy loss of language models follows a power-law relationship with the number of parameters N, dataset size D, and compute C, approximately L ∝ N^(-0.076), L ∝ D^(-0.095), L ∝ C^(-0.050).75% similarVerifiedCross-entropy loss originates from KL divergence in information theory and measures the degree of difference between the true probability distribution and the predicted probability distribution.73% similarUnverifiedCross-entropy loss often converges faster than MSE in classification tasks because, when paired with softmax output, gradients do not vanish in the saturation region.71% similarUnverified标准交叉熵损失在目标占图像面积较小时会导致严重的类别不平衡问题，模型倾向于全部预测为背景69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/810525API
curl https://kongchang.com/api/v1/knowledge/claims/810525MCP
get_claim(id=810525)