Unverified50% confidenceFactExact time
AUC is the standard metric for measuring the performance of a binary classifier, with values from 0.5 to 1.0, and 98.8% AUC means that in about 99% of sample pairs, the probe can correctly rank conversations containing deception ahead of conversations without deception.
1
Sources
50%
Confidence
Long-term
Relevance
10/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn introspection-style deception tests, the probe needs to distinguish conversation records containing the model's true hidden objective from records containing other objectives, and the probe's AUC reaches as high as 99.7%.78% similarUnverifiedAUC是衡量二分类模型综合性能的核心指标，0.5代表随机猜测，1.0代表完美分类77% similarUnverifiedAUC将所有阈值下的表现等权平均，掩盖了在特定低误报率工作阈值下的真实表现，在高安全场景可能给出误导性评估74% similarUnverifiedAUC is a comprehensive evaluation metric for binary classification models, with a value range of 0.5 to 1.0; it is not affected by class imbalance, and in clinical prediction, an AUC of 0.7–0.8 has some reference value, 0.8–0.9 is considered good, and above 0.9 is considered excellent.73% 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.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1013474API
curl https://kongchang.com/api/v1/knowledge/claims/1013474MCP
get_claim(id=1013474)