Unverified50% confidenceFactExact time
Lexical coverage measurements showed: exact match 1.00, paraphrase 0.44, third-party question 0.63; with the threshold set at 0.9, queries that genuinely need semantics open the semantic channel.
1
Sources
50%
Confidence
Long-term
Relevance
10/18/2026
First Seen
Sources
Related Entities
Related Claims
Unverified通义千问3.6 27B在语义三次全对率和格式正确率上均领先Muse Glimmer70% similarUnverifiedOn the 35 prompts that the router kept local, Llama 3.2 1B's answers were judged acceptable 30 times (86%, average 4.4/5); on the 23 text prompts sent to the cloud, the same model was acceptable only 4 times (17%, average 2.0/5).69% similarUnverifiedRelevance gating can intercept clearly off-topic questions (e.g., toasting bread scores 0.25, below the 0.45 threshold), but borderline questions score 0.54 to 0.63, actually higher than several questions the blog can answer; intercepting them all would wrongly reject 13 of the 30 real questions.69% similarUnverifiedIn an evaluation of 30 answerable questions, off-the-shelf small models under 350MB achieved faithfulness (all assertions can be found in the passages) at most 3 times.68% similarUnverifiedTesting the query prefix mentioned in the bge model documentation, "Represent this sentence for searching relevant passages:", the top score improved by 0.014, but the gap to the second place actually narrowed.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1062256API
curl https://kongchang.com/api/v1/knowledge/claims/1062256MCP
get_claim(id=1062256)