Unverified50% confidenceBenchmarkExact time
The model judges' agreement with human adjudication labels on 1,860 elements (Gwet's AC1) was at least 0.894.
1
Sources
50%
Confidence
Long-term
Relevance
10/22/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedPew's ATP design effect typically ranges between 1.5 and 2.5, meaning that a survey with a nominal sample size of 1,000 may have an effective sample size of only 400–667.71% similarUnverifiedOn the internal test set (11,773 AI-involved files and 2,541 human files), the method achieved a macro F1 of 72.4 at the 1% false positive rate operating point.70% similarUnverifiedIn the Phase 1 post-hoc comparison, SAGE's evaluation results were compared against 105 human reviews from the original competition, yielding an ordinal consistency of kappa = 0.29, which falls in the "fair" range.68% similarUnverifiedOn a test set of 600 synthetic invoices, the QLoRA fine-tuned model achieved a full-invoice exact rate of 83.0%, an auto-approval rate of 86.8%, and 0 out of 461 validation-coverage field errors in approved invoices (upper bound 0.8%).67% similarUnverifiedOn synthetic data with 10,000 samples, a 1% positive rate, and 495 negatives scoring 0.90 while the positives score 0.80 and 0.60 respectively, computing with a 0.5 threshold gives: ROC-AUC=0.9500, Average Precision=0.1449, precision=0.1681, recall=1.0000, accuracy=0.9505, TP=100, FP=495, TN=9405, FN=0.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1084335API
curl https://kongchang.com/api/v1/knowledge/claims/1084335MCP
get_claim(id=1084335)