Evalica, your favourite evaluation toolkit

route and climber ratings for rock climbing

Source code and data for the EDM 2022 paper

Package to do Bradley-Terry Model pairwise compairsons

Concept-Guided Chain-of-Thought (CGCoT) pairwise annotation tool for systematic text evaluation using LLMs. Generate breakdowns, compare items, compute scores, and validate against human judgments. Supports Ollama, Hugging Face, Google Gemini, OpenAI, and Anthropic models.

Moody Lenses internal campaign decision engine — LLM-powered multi-dimensional evaluation for marketing campaign selection

Extract archery recurve and compound event scores from Ianseo and builds a website containing the resulting ranks of all archers. The statistical analysis was written in R, using the PlackettLuce package.

public ranking system for ai agents

AI-powered personal recommendation engine that learns preferences through pairwise comparison

R Package: Pairwise Comparison Tools for LLM-Based Writing Evaluation

Code for the paper 'Bias-Aware Ranking from Pairwise Comparisons' (BARP)

UI for straightforward Bradley-Terry feedback loop

Self-hosted blind LLM evaluation: build a campaign, collect human or simulated votes, ship a defensible Bradley-Terry ranking.

Oscars predictions using Bradley-Terry model and stan

An LLM benchmark where models create memes from current news headlines and humans vote on the results.

This is the online appendix for the paper "Bayesian Paired-Comparison with the bpcs package"

Fast inference for multi-body rankings using Newman's efficient Plackett-Luce algorithm. Achieves 5-70x speedup with Numba JIT compilation.

Bayesian Spatial Bradley--Terry

Bayesian player ability model for predicting golf win probabilities — time-varying ratings, Bradley-Terry/Glicko, historical PGA Tour data

TypeScript CLI + MCP server implementing OpenDeepThink (Zhou et al., arXiv:2605.15177). Independent reimplementation; not a fork.