Unverified50% confidenceSolutionExact time
Rather than arguing about which metric has a higher average score, it is better to output the list of affected queries, including: which queries are judged differently under the two rules, the expected evidence for each query, the actual retrieval position, and the specific rule that causes the conclusion to flip.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen only querying without needing to modify files, the Ask mode should be preferred because it consumes fewer credits and offers higher safety; using Craft for queries consumes more credits.75% similarUnverifiedThe evaluation set should cover several types of edge cases: rephrased questions, questions requiring combining two facts, queries containing spelling errors, and questions that no source can answer.74% similarUnverifiedThe retrieval layer and generation layer should be evaluated separately; otherwise, a single answer quality score cannot indicate where to investigate.73% similarUnverifiedThe accuracy of query results from NL2SQL tools depends on the quality and governance of the underlying data, and vague or ambiguous questions may yield misleading answers.70% similarUnverifiedSemantic caching bypasses the model call entirely when a new query is sufficiently similar to an already-answered query, but carries the trade-off of returning incorrect answers to subtly different questions.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1048190API
curl https://kongchang.com/api/v1/knowledge/claims/1048190MCP
get_claim(id=1048190)