benchmark source-backed
Jev reranking: saved outputs across retrieval tasks and scoring strategies
A retrieval study compares typed Jev scoring strategies with dedicated rerankers and open-model controls on the same passage candidates. Saved provider responses, query-level evidence, bootstrap intervals and an interactive viewer support inspecting each result.
Notes
On eight English datasets and 1,617 scored questions, the author reports equal-dataset nDCG@10 of 0.692 for Jev's four-level rubric and 0.691 for Cohere Rerank 4 Pro; equal-query weighting gives a different comparison.
source-backed — Public repository, docs or live artifact. About this label
Destinations
Try a walkthrough
- Select relevant context with Jev →
Score a small shortlist of passages, preserve their source IDs and pass selected evidence to your answering model.
Verification levels
- HTTP availability
- Link reachable ·
Method: bounded HTTP request with redirects. HTTP 200 · HTTP response only. Checked destination ↗ · Last reachable - Browser demo scenario
- Not checked
- Build
- Not checked
- Live API test
- Not checked