Unverified50% confidenceSolutionExact time
The two-stage design of Key routing is "coarse routing + fine scoring", that is, first predicting at low cost which Key pages are most relevant to the current Query, and then performing exact QK dot-product computation only on those pages.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedClueFlow was evaluated on four search benchmarks, covering both live web and fixed corpus settings.63% similarUnverifiedThe key selection algorithm first prioritizes keys with a request cap that still have remaining quota (with more remaining quota preferred), then keys with no cap, in order to use free keys first with paid keys as a fallback.63% similarUnverifiedQKV分别代表Query（查询）、Key（键）和Value（值），命名借鉴自信息检索系统概念61% similarUnverifiedCommon implementations of tiered routing include rule-based length/perplexity thresholds, using small classifiers to predict query complexity scores, and cascade routing (where a weak model attempts first and then decides whether to escalate).60% similarUnverifiedFTS5 is based on the inverted index principle, recording the documents and positions where each term appears using terms as keys, which is the same underlying principle as search engines.60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1130479API
curl https://kongchang.com/api/v1/knowledge/claims/1130479MCP
get_claim(id=1130479)