Unverified50% confidenceBenchmarkExact time
At 66.2% coverage, the selective macro F1 score was 99.01%.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
Unverified99.3%通过率（283/285）对应的95%置信区间约为97.5%-99.9%（使用Wilson区间估计）73% similarUnverifiedA sparse subset comprising only 28%–48% of upstream components is sufficient to retain 88%–101% of the full steering effect.71% similarUnverifiedAt τ=0.6 the confidence distribution is concentrated around a median of 0.6294, and 851 of the 2000 decisions (42.6%) fall below 0.6, which by case count is 90.8%.70% similarUnverifiedIn the control group where the answer key was removed from the prompt, the match rate plummeted from 63% to 1% (1/75).70% similarUnverifiedResearch shows that inter-annotator agreement rates for RLHF are typically only between 60% and 75%.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073958API
curl https://kongchang.com/api/v1/knowledge/claims/1073958MCP
get_claim(id=1073958)