Unverified50% confidenceFactExact time
Permutation importance measures importance by randomly shuffling a feature's values and observing the drop in model performance; it is simple to compute but may underestimate the importance of correlated features.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDenser relevance judgments reduce the probability that correct results retrieved by a model are mistakenly judged as irrelevant, thereby improving evaluation reliability.76% similarUnverifiedThe impact of data quality on fine-tuning results is often far greater than that of hyperparameter tuning, and this is especially critical in scenarios with limited sample sizes.76% similarUnverifiedIgnoring error correlation causes the combined variance to be underestimated, and the underestimation factor grows with group size; this factor accurately predicts the degree of coverage failure.75% similarUnverified权重初始化的随机性对模型性能影响最深远，数据批次打乱顺序的随机性影响最小75% similarUnverifiedRandom error can be averaged out by increasing sample size, but systematic bias is amplified with sample size, causing attack success rates to be structurally overestimated or underestimated.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1074558API
curl https://kongchang.com/api/v1/knowledge/claims/1074558MCP
get_claim(id=1074558)