Unverified50% confidenceFactExact time
The composite score in this study places 80% weight on the large language model's judgment of factual entailment and 20% on lexical overlap.
1
Sources
50%
Confidence
Long-term
Relevance
9/15/2026
First Seen
Sources
Related Entities
Related Claims
Unverified80% of accuracy improvement comes from engineering, and 20% comes from model capability.71% similarUnverified2023年学术界研究发现，顶级大模型作为评判者与人类专家的判断一致性可达80%以上70% similarUnverified功效分析中统计功效（Statistical Power）通常设为 80% 以上70% similarUnverifiedIn a well-calibrated model, an output score of 0.8 should statistically correspond to an actual accuracy of 80%.69% similarUnverified学术研究已证明，同样的问题以不同方式表达，模型的回答准确率可能相差20%以上69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/921808API
curl https://kongchang.com/api/v1/knowledge/claims/921808MCP
get_claim(id=921808)