Human-preference ranking app for comparing LLM responses with pairwise voting and ELO-style scoring.
-
Updated
May 2, 2026 - JavaScript
Human-preference ranking app for comparing LLM responses with pairwise voting and ELO-style scoring.
Python software for calculating distances between preference rankings, including ties.
Structured frontend fidelity evaluation — two AI-generated UI replications judged against a reference interface, with DevTools inspection, severity-ranked findings, and an evidence-based preference decision.
Sorting made better, powered by science.
To associate your repository with the preference-ranking topic, visit your repo's landing page and select "manage topics."