Unverified50% confidenceFactExact time
A false positive is when a test reports failure but there is actually no bug, and a false negative is when a test reports pass but a bug actually exists.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedResumate's failure detection is exception-based and cannot detect "wrong but valid" outputs that contain incorrect content without throwing an exception.65% similarUnverifiedWEAK tests are tests that exist but whose failure reason does not point to the behavior under test, for example a test file that fails to load due to importing a nonexistent name, dying at the collection or import stage rather than at the assertion stage.65% similarUnverifiedThe root cause why the check could never fail was that the prompt itself already contained a constraint that made it impossible for the defect being checked to be generated in the first place.60% similarUnverifiedCalibration measures whether the confidence reported by the model matches actual accuracy; an unvalidated confidence interface may mislead downstream decisions and should not be shipped.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1048112API
curl https://kongchang.com/api/v1/knowledge/claims/1048112MCP
get_claim(id=1048112)