Unverified60% confidenceFactExact time
When two unit vectors are randomly sampled in d-dimensional space, their inner product has an expected value of 0 and a standard deviation of approximately 1/√d; when d=1024, the standard deviation is approximately 0.031.
2
Sources
60%
Confidence
Long-term
Relevance
8/8/2026
First Seen
Sources
Related Claims
Unverified高维空间中随着维度增加，随机向量之间的夹角趋向集中在90°附近，极端值1或-1概率趋近于零73% similarUnverifiedIn the fixed-vector self-similarity experiment, when the offset goes from 0 to 1024, the mean of the raw scores always fluctuates within ±0.38, while the mean of the absolute values only drops from 3.17 to 3.02, a change of about 4%.68% similarUnverifiedThe binary threshold design is unfavorable to models near the threshold, as 0.74 and 0.80 differ drastically in score even though the actual capability gap may be negligible.65% similarUnverifiedWith a sample size of 112 questions, the 95% confidence interval near 0.95 is approximately ±4 points, making it impossible to distinguish between scores of 0.94 and 0.92.64% similarUnverifiedUsing Jensen-Shannon divergence between time-domain and frequency-domain models as an uncertainty measure, AURC was 0.29 on the standard split (better than 0.36) and 0.54 under single-condition reproduction (better than 0.73).63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/705786API
curl https://kongchang.com/api/v1/knowledge/claims/705786MCP
get_claim(id=705786)