Unverified50% confidenceBenchmarkExact time
Using only 64 paired segments, raw Recall@1 can be raised to 88.4%–90.3%, and token cosine similarity to about 0.82.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe experimental results show that even with identical inputs, the raw Recall@1 between latent representations produced by different schemas is only 5.9%–17.7%.71% similarUnverifiedIn the composite score for memory retrieval, similarity has the largest weight, recency accounts for a small proportion, and the result is multiplied by an importance factor with a range of approximately 0.8 to 1.270% similarUnverified语义缓存基于向量嵌入的余弦相似度，实践中通常将0.92-0.95设为命中阈值65% similarUnverifiedOn the hard tier, recall@5 improved from 0.000 to 0.955 (under default blending weights), and reached 1.000 when using the lower-bound variant.64% similarUnverifiedIn the control group where the answer key was removed from the prompt, the match rate plummeted from 63% to 1% (1/75).63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1023827API
curl https://kongchang.com/api/v1/knowledge/claims/1023827MCP
get_claim(id=1023827)