Unverified50% confidenceBenchmarkExact time
The confidence reported by the model is severely disconnected from actual accuracy: the 95% confidence tier had an actual accuracy of 0%, while the 90% and 70% tiers both had an actual accuracy of 100%.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedA gating accuracy of 58% reported in the paper was mistakenly treated as end-to-end quality, whereas the true end-to-end quality is 98%.80% similarUnverifiedResearch shows that when humans assign 90% confidence, their actual accuracy rate is often only around 70%, indicating a systematic overconfidence80% similarUnverifiedDifferent papers often use faithfulness threshold values ranging from 70% to 95%, lacking a unified standard.78% similarUnverifiedWhen Jev returns a confidence level above 0.95 (approximately 60% of messages), accuracy reaches 95.8%, indicating reliable confidence calibration quality78% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1005671API
curl https://kongchang.com/api/v1/knowledge/claims/1005671MCP
get_claim(id=1005671)