Unverified50% confidenceBenchmarkExact time
The best-performing model reached 0.805 on strict balanced accuracy, with a tBA of 0.855.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
11/17/2026
First Seen
Valid until: 12/1/2026
Sources
Related Entities
Related Claims
UnverifiedThe framework achieved a precision of 0.82, a recall of 0.85, and an F1 score of 0.82.77% similarUnverifiedAfter introducing the log-odds score, precision rose to 0.8580 and AUC rose to 0.852.76% similarVerifiedIn a well-calibrated model, an output score of 0.8 should statistically correspond to an actual accuracy of 80%.74% similarUnverified研究表明经过良好校准的LLM裁判与人工标注的一致性可达80%以上71% similarUnverifiedOn action appropriateness, the average score across all models reached 95.6%.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1179329API
curl https://kongchang.com/api/v1/knowledge/claims/1179329MCP
get_claim(id=1179329)