Unverified50% confidenceFactExact time
Most evaluation metrics such as precision and F1 cannot distinguish between a system that checked and found no relevant evidence and a system that did not check at all.
1
Sources
50%
Confidence
Long-term
Relevance
9/15/2026
First Seen
Sources
Related Entities
Related Claims
Unverified验证集和测试集不应做任何数据增强，评估应在真实未经处理的数据上进行70% similarUnverifiedMost commercial detectors do not disclose real test results of their accuracy rates, and the project still lacks a verifiable complete dataset and methodology documentation.68% similarUnverifiedRecording only the proposed calls and tool responses is insufficient to determine the true cause of an agent security test result.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/922256API
curl https://kongchang.com/api/v1/knowledge/claims/922256MCP
get_claim(id=922256)