QwenPaw tool plugin: multi-judge consensus ranking — cross-family LLM judges each rank an independently anonymized slate; mean-rank consensus, Spearman report, per-judge position-stability.
python spearman-correlation llm-evaluation llm-as-a-judge qwenpaw qwenpaw-plugin listwise-ranking consensus-ranking
-
Updated
Oct 9, 2026 - Python