Unverified50% confidenceDecision RuleExact time
In distance-bucketed evaluation, it is generally recommended to have no fewer than 50–100 valid samples per bucket for the accuracy estimate in that interval to have practical reference value.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedEach question in the test is scored out of 100 points, consisting of 70 points for correctness plus 30 points for quality, with different questions carrying different weights.69% 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).69% similarUnverifiedThe pLDDT score ranges from 0 to 100; scores above 90 are generally considered high-confidence predictions, 70–90 are considered good predictions, and scores below 50 should be interpreted with caution.67% similarUnverified在5%显著性水平下测试100个随机因子，期望会有5个看起来显著（多重假设检验问题）67% similarVerifiedExpected Calibration Error (ECE) bins prediction samples by confidence level, computes the weighted average of the difference between predicted confidence and actual accuracy within each bin, and ideally approaches 0.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/983665API
curl https://kongchang.com/api/v1/knowledge/claims/983665MCP
get_claim(id=983665)