Expired50% confidenceBenchmarkExact time
After fixing TOP_K to 8, the top score is 0.635, the second row is 0.592, and the language/education chunk at the end of the list scores below 0.45, with a visible separation reappearing at the top.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/17/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/31/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedReciprocal Rank@k is defined as the reciprocal of the rank of the first relevant result: ranking first gives 1 point, ranking third gives 1/3 point, and having none at all gives 0 points.64% similarUnverifiedTop-K retrieval refers to returning the top K chunks with the highest similarity during the retrieval stage, and K is usually taken from 3 to 10.64% similarUnverifiedOn real storage, the top score is 0.635 and the tail is only 0.450; in the 0.185 gap only the first row deserves attention, and everything below 0.55 is a noise band.64% similarVerifiedEmbeddingGemma 2's score on the MTEB Code benchmark improved from 68.76 in the first generation to 78.68, an increase of 9.92 points.63% similarUnverifiedThe author tested the query "Kubernetes, Terraform, SRE on-call" against a 6-chunk demo resume, and the relevant chunk scored 0.784 while the remaining chunks scored between 0.573 and 0.487.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1054549API
curl https://kongchang.com/api/v1/knowledge/claims/1054549MCP
get_claim(id=1054549)