Unverified50% confidenceFactExact time
The model uses binary cross-entropy loss and is trained end-to-end against oracle extractive labels.
1
Sources
50%
Confidence
Long-term
Relevance
9/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe binary cross-entropy loss formula is -[y·log(p) + (1-y)·log(1-p)], where y is the true label and p is the model's predicted probability for the positive class.73% similarUnverifiedCross-entropy loss is the most commonly used loss function in classification tasks, rooted in information theory, and measures the distance between the model's predicted probability distribution and the true distribution.70% similarUnverified交叉熵恢复率通过将SAE重建激活替代原始激活输入到模型后续层比较交叉熵损失变化来计算，衡量SAE是否保留对下游计算重要的信息68% similarUnverifiedThe objective function for pre-training is typically the cross-entropy loss in causal language modeling, i.e., maximizing the probability of correctly predicting the next token given all preceding tokens.67% similarUnverified交叉熵损失函数鼓励模型将正确类别的logit推向无穷大，导致深度网络预测过于确定65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/921877API
curl https://kongchang.com/api/v1/knowledge/claims/921877MCP
get_claim(id=921877)