Unverified50% confidenceFactExact time
In credit risk scenarios, if training samples include repayment records updated after the application occurred, the model's AUC may be artificially inflated by 5 to 10 percentage points, with performance dropping sharply after deployment
1
Sources
50%
Confidence
Long-term
Relevance
9/24/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen the ratio of positive to negative samples is extreme, the area under the PR curve (AUPRC) is often more revealing of true performance than AUC.69% similarUnverified即使只有5-10%的训练样本存在错位，也会导致VLM模型在下游任务上出现显著的性能退化68% similarUnverified当训练窗口、门控窗口和报告窗口未严格分离时，PR-AUC达到0.72到0.76，严格分离后骤降至约0.51接近随机水平，这是数据泄漏的表现68% similarUnverifiedIn the intervention experiment, the donor-control AUC rose from 0.254 to 0.540, with a paired change of +0.286 (95% confidence interval [+0.272, +0.300]), and an increase was observed across all 18 concepts.67% similarUnverified研究表明当ReAct循环超过10步时，模型的任务完成率会出现显著下降65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/947808API
curl https://kongchang.com/api/v1/knowledge/claims/947808MCP
get_claim(id=947808)