Unverified50% confidenceOpinionExact time
Aggregate scores mask imbalance across dimensions, so interpretation should combine per-metric indicators to judge weaknesses.
1
Sources
50%
Confidence
Long-term
Relevance
11/17/2026
First Seen
Sources
Related Entities
Related Claims
Unverified聚合多个基准的评测方式降低了单一基准被针对性优化(刷榜)的风险77% similarUnverifiedCommonly used aggregate metrics (such as overall accuracy and evidence coverage) miss claim-level patterns, and macro-level numbers can become disconnected from actual per-claim changes.76% similarUnverifiedAggregate metrics naturally dilute concentrated failure patterns, making localized failures invisible within the overall number75% similarUnverifiedExact match is a strict lower bound that underestimates the actual usable prompt quality; graded metrics and LlamaPIE's built-in judge correlate more highly with human ratings than exact match.75% similarUnverifiedIn class-imbalanced scenarios, accuracy alone can be misleading and needs to be combined with a confusion matrix for more fine-grained evaluation.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1179641API
curl https://kongchang.com/api/v1/knowledge/claims/1179641MCP
get_claim(id=1179641)