Unverified50% confidenceFactExact time
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
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
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72% 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.71% similarVerifiedCross-Entropy Loss is commonly used for classification tasks, and Mean Squared Error (MSE) is commonly used for regression tasks.69% similarUnverifiedToolformer的因果筛选机制仅当API调用结果使模型对后续token的交叉熵损失降低超过阈值时，该样本才进入微调数据集68% similarUnverifiedLLM的幻觉问题源于其训练目标是最小化交叉熵损失而非优化命题真值，是与生成机制深度耦合的固有特性而非偶发缺陷68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/527098API
curl https://kongchang.com/api/v1/knowledge/claims/527098MCP
get_claim(id=527098)