Unverified50% confidenceOpinionExact time
When monitoring the error rate of large model APIs, focusing only on the overall success rate is insufficient; attention should be paid to error rate percentiles and spike detection, because averages may mask the impact of tail error rates on some users.
1
Sources
50%
Confidence
Long-term
Relevance
10/24/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe error rate of manual data entry is typically between 0.5% and 4%, and the error rate is higher for tasks involving complex judgment.79% similarUnverifiedIn recommendation scenarios, models typically use Root Mean Square Error (RMSE) to measure the error between predicted ratings and actual ratings.77% similarUnverifiedThe oracle router knows in advance which cases the small model will get wrong, sends only the wrong requests to the large model, and achieves accuracy comparable to always using the large model at only 29% of the cost.77% similarUnverified工业SCADA故障事件数据高度不平衡，故障事件通常占比不足1%，普通分类模型会偏向正常状态导致低召回率76% similarUnverifiedRecording negative samples where 'a recommended solution did not apply' is just as important as recording successes; otherwise the model will fall into statistical bias, mistaking high-frequency solutions for high-accuracy ones76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1092145API
curl https://kongchang.com/api/v1/knowledge/claims/1092145MCP
get_claim(id=1092145)