Unverified50% confidenceFactExact time
Sample-level leakage refers to the situation where, when multiple augmented samples originate from the same original sample, if the split is done randomly rather than by original individual, information about the same entity appears in both the training set and the validation set.
1
Sources
50%
Confidence
Long-term
Relevance
9/30/2026
First Seen
Sources
Related Entities
Related Claims
Unverified预处理泄漏在小数据集上尤为显著，当训练集仅有数百张图像时加入测试集样本会明显改变均值和标准差估计77% similarUnverifiedPerforming StandardScaler normalization on the entire dataset before the train/test split causes the mean and variance information of the test set to leak into the training set, constituting data leakage.76% similarUnverifiedData leakage falls into two categories—target leakage and training set leakage—and causes models to perform well during testing but fail after deployment.73% similarUnverified冻结基准可能因数据泄露被纳入模型训练数据而随时间失去区分力，这是一个权衡71% similarVerifiedBenchmark data contamination refers to the phenomenon where, after a test set is made public, its questions may appear in the pre-training corpora of subsequent models, causing scores to reflect memorization rather than genuine reasoning ability.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/961894API
curl https://kongchang.com/api/v1/knowledge/claims/961894MCP
get_claim(id=961894)