Unverified50% confidenceBenchmarkExact time
The Rejection Sampling method achieves an accuracy of 0.797 and a span F1 of 0.556, with verbatim citation improving from 0.597 to 0.701.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAfter removing the three leakage sources—metadata, news agency labels, and duplicate documents—F1 drops only from 0.9935 to 0.9814, a decrease of 1.21 percentage points.67% similarUnverifiedCSP improves the F1 score for rule consistency determination from 0.90 to 0.94.67% similarUnverifiedIn the study, the Rowwise method achieved an SHD of 160 and an F1-score of 0.77.66% similarUnverifiedFine-tuned DistilBERT achieves F1=0.9993 on in-distribution testing, but loses 12.9 mean average precision points under topic shift, a greater degradation than the linear model's 5.2 points.66% similarUnverifiedEvidence-driven pre-improvement validation improved precision from 58.6% to 72.3%, and reduced unsupported objectives per round from 1.93 to 0.50.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/982999API
curl https://kongchang.com/api/v1/knowledge/claims/982999MCP
get_claim(id=982999)