Unverified50% confidenceFactExact time
研究表明人类评估者在判断LLM输出的事实准确性时，正确率往往只有50%-70%
1
Sources
50%
Confidence
Long-term
Relevance
8/27/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn financial statements, character accuracy may be as high as 99%, but semantic accuracy may be less than 50%.78% similarUnverifiedResearch shows that the agreement rate between LLM judges and human scores is approximately 80%, approaching but not yet surpassing the level of human inter-annotator agreement (approximately 81%).78% similarUnverified在事实性评估任务上模型评判者准确率仍显著低于人类专家，但在风格、连贯性等主观维度已接近人类评审员间一致性水平76% similarUnverifiedResearch shows that when humans assign 90% confidence, their actual accuracy rate is often only around 70%, indicating a systematic overconfidence74% similarUnverified泰特洛克的研究得出结论：专家预测的准确率仅略高于随机猜测，甚至不如简单的统计基准线74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/810185API
curl https://kongchang.com/api/v1/knowledge/claims/810185MCP
get_claim(id=810185)