Verified65% confidenceBenchmarkExact time
In a reproduction experiment training for 40 epochs on CIFAR-10, plain-56 achieved a training accuracy of 84.0% and plain-20 achieved a training accuracy of 95.1%, even though plain-56's parameter count (853,018) is approximately three times that of plain-20 (269,722).
3
Sources
65%
Confidence
Long-term
Relevance
8/15/2026
First Seen
Sources
深度网络退化问题实验复现：残差连接为何能拯救深层网络
Related Claims
UnverifiedIn the same reproduction experiment, ResNet-20 achieved a test accuracy of 90.4% and ResNet-56 achieved a test accuracy of 91.7%, with training accuracies of 97.4% and 99.0% respectively.79% similarUnverifiedDuring training, the held-out FID dropped from 33.7 to 27.0, FD-DINOv2 dropped from 570 to 218, and detector-based object accuracy improved from 65% to 90%.73% similarUnverifiedAfter 15 training epochs on the demonstration dataset, the text classification model achieves an accuracy of 99%, with a total of 110,000 model parameters (16,000 for Attention, 33,000 for FFN, and the remainder for Embedding)70% similarVerified如果一个10步操作中每步准确率为95%，整体成功率仅约60%70% similarUnverifiedThe handwritten MLP model can achieve approximately 98.5% accuracy on the full MNIST training set.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/750220API
curl https://kongchang.com/api/v1/knowledge/claims/750220MCP
get_claim(id=750220)