Unverified50% confidenceFactExact time
A study published on arXiv (arXiv:2609.02899) proposes that concerns about benchmark contamination should be distinguished into two separate questions: whether absolute scores are inflated and whether the ranking order of models is changed.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
Unverified研究表明即使是部分污染（模型只见过题目而非答案），也可能显著提升得分66% similarPartially VerifiedBenchmark contamination refers to a model having already seen test questions during training, resulting in inflated scores that fail to reflect true generalization ability.63% similarUnverifiedOnly differential contamination—where some models are contaminated far more than others—distorts rankings, and this situation is rare in practice.62% similarUnverifiedRecent research indicates that some models perform 10–15 percentage points higher on known contaminated benchmark subsets than on uncontaminated subsets.62% similarUnverified评估模型不能只看智能指标，还要看Token使用效率和工具调用次数，工具调用次数过多会导致上下文污染61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/900504API
curl https://kongchang.com/api/v1/knowledge/claims/900504MCP
get_claim(id=900504)