Unverified50% confidenceFactExact time
Computing p95 from twelve samples actually lands between the 11th and 12th samples—essentially the second-slowest run—and is extremely unstable.
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedTo discuss p95 you need at least dozens of samples, and p99 requires hundreds; twelve runs can only provide existence evidence (missing usage field, 429, whether the tail is two seconds or twenty) and cannot precisely estimate quantiles.73% similarUnverifiedPerformance benchmark results are subject to statistical fluctuation, requiring multiple runs to take the median or P95/P99 percentiles for evaluation.71% similarUnverified若单一模型漏检率为 p，两个训练背景不同的模型同时漏检同一错误的概率近似降至 p²，20%漏检率理论上可降至 4%70% similarVerifiedReAct's errors amplify in a chain: if the accuracy per step is p, the overall success rate after n steps is approximately p to the power of n; with a per-step accuracy of 95%, the overall success rate drops to approximately 59.9% after 10 steps and only 35.8% after 20 steps.69% similarUnverified若单帧误检率为p，连续8帧独立误检的概率为p^8，即使单帧误检率高达10%，8帧连续误检概率也只有10^-868% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1067315API
curl https://kongchang.com/api/v1/knowledge/claims/1067315MCP
get_claim(id=1067315)