Unverified50% confidenceBenchmarkExact time
13 matched document pairs are enough for the paired test to reach 80% statistical power at the 5% significance level, for detecting whether a certain set was trained on.
1
Sources
50%
Confidence
Long-term
Relevance
11/2/2026
First Seen
Sources
Related Entities
Related Claims
Unverified实验采用8:2比例划分数据，20%作为内部测试集，训练集内部使用5折交叉验证，并移除重复记录以避免数据泄漏77% similarUnverifiedThe test case provided in the course uses inputs of 80 and 92 to verify whether the result correctly reflects an increase of 15 percentage points.75% similarUnverifiedThe accuracy data in the test has an error margin of approximately ±8–10 percentage points at n=100 sample size (based on a 95% confidence-level binomial confidence interval).72% similarUnverifiedThe test case generation solution is divided into five progressive levels: a single agent plus prompt achieves only approximately 30% coverage, a workflow approach can achieve approximately 80% coverage, and an exploratory learning approach can achieve over 90% coverage.72% similarUnverifiedA single "trusted source" prompt caused 7 out of 8 tested models to reverse between 45% and 88% of their correct answers.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1125009API
curl https://kongchang.com/api/v1/knowledge/claims/1125009MCP
get_claim(id=1125009)