jev / gallery
← Back to collection

Author test

jev-rerank-bench

Test Jev as a search reranker

How this project uses Jev

Input
Candidate passages returned by BM25
Jev decides
Which passages are relevant and their order
Code executes
Evaluation code computes retrieval metrics and exposes raw output

Evidence and limitations

The repository compares multiple datasets. Different averaging methods can change rankings; small average differences do not establish an overall winner.

This project has not been run independently here. Author-reported results are not independently verified results.

Original sources

Explore related use cases

Browse this category →