Unverified50% confidenceFactExact time
Judge disagreement predicts errors, with AUROC up to 0.80, but the variance inflation from inverse probability weighting exceeds the signal gain.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedUsing the behavioral features of each state for failure prediction, the AUROC on the held-out set reaches as high as 0.94.77% similarUnverifiedWhen 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.76% similarUnverifiedIn rare-positive scenarios (such as fraud, abuse, and defect detection), AUC may look high, but because the negative pool is enormous, even a low false-positive rate produces many false alarms, resulting in very low precision.73% similarUnverifiedCalibration means that when the model outputs 0.9, across a large number of similar samples about 90% are indeed the correct class; an uncalibrated model can have high AUC but its values cannot be used directly as probabilities.70% similarUnverifiedThe Expected Calibration Error (ECE) increased in seven out of eight event pairs, and in affected scenarios, 12%–51% of the area was confidently mislabeled by the model with near-zero accuracy.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1069907API
curl https://kongchang.com/api/v1/knowledge/claims/1069907MCP
get_claim(id=1069907)