Unverified50% confidenceFactExact time
Out of a total of 188 model × benchmark combinations, only 3 cases exhibit differential contamination cross-validated by two reference points.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOnly differential contamination—where some models are contaminated far more than others—distorts rankings, and this situation is rare in practice.69% similarUnverifiedRecent research indicates that some models perform 10–15 percentage points higher on known contaminated benchmark subsets than on uncontaminated subsets.63% similarUnverified该测试样本规模相对有限（5个模型、13个代码库），结论有待更大规模验证59% similarUnverified多重比较问题指同时测试多个模型变体时，即使每个单独检验显著性水平设为0.05，整体假阳性率也会急剧膨胀58% similarVerifiedIn scenarios with extreme class imbalance, even an accuracy as high as 95% may indicate that the model is completely unable to identify the minority class.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/900512API
curl https://kongchang.com/api/v1/knowledge/claims/900512MCP
get_claim(id=900512)