llm-rankers: checking Jev's calibration with TREC judgments

The ielab document-ranking toolkit gained a jev module, with code that uses TREC human relevance judgments to test whether Jev's probabilities are actually calibrated.

Notes

Code for the calibration check; results in the repository.

source-backed — Public repository, docs or live artifact. About this label

Destinations

Filed under

• #ranking
• #calibration
• #research
• #benchmark

In collections

• Benchmarks & independent tests →
• Search, ranking & queues →