Unverified50% confidenceFactExact time
Among the study's four retriever configurations, the selected hybrid retriever with a 384-token chunk size and 20% overlap performed best, retrieving accepted evidence in the top 3 results for 17 of the 18 queries.
1
Sources
50%
Confidence
Long-term
Relevance
10/23/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn a self-built test set (120 real queries, 38 of which contain identifiers), pure vector retrieval hit 21/38 identifier queries, pure BM25 (with an identifier-preserving tokenizer) hit 36/38, and hybrid + RRF hit all 38.66% similarUnverifiedOn the 14-clause evaluation set, gemma3:4b achieved 50% classification accuracy (13 of 14 clauses were judged yes) and 100% verbatim excerpt accuracy (all 14 quotes matched verbatim).64% similarUnverifiedIn a small-scale test published by the PartsPlan developer, all 30 chart values were answered correctly.64% similarUnverifiedOn a benchmark of 30 questions across five categories, graph retrieval scored 24.667 and flat RAG scored 19.251.64% similarUnverifiedIn an internal organizational survey, the VERA dashboard received a usefulness rating of 4.36 out of 5.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1088185API
curl https://kongchang.com/api/v1/knowledge/claims/1088185MCP
get_claim(id=1088185)