Orwell is an open-source, local-first platform for systematically auditing and evaluating Large Language Models (LLMs) for bias, safety violations, and behavioural blind spots — before they reach production.
EquiLens is an AI fairness platform that detects and mitigates bias using statistical metrics, ML models, and Google Gemini’s contextual insights. It offers semantic feature analysis, computes metrics like Demographic Parity and chi-square, and applies automated bias reduction. It also generates PDF reports, enabling users to build fairer models
Lean, research-backed multi-LLM council with bias-aware judge controls (position-swap, verbosity-norm, stake-free judge) adapted from LLM-eval research.
A small and simple prototype designed to alert users of the bias of the news source.
This German scientific poster (developed as HTML5/React web app) shows the potential implications of racial/social/... Biases in the context of Machine Learning
E-learning website teaching users to identify logical fallacies, built with React.
A fairness-aware loan approval system that detects, analyzes, and mitigates bias in machine learning models using fairness metrics and human-in-the-loop decision making.
Official academic page for the paper: "Diffusing DeBias: Synthetic Bias Amplification for Model Debiasing" (NeurIPS 2025)
Five-minute timer, checklist and report for quick recorded expert website reviews.
To associate your repository with the bias-mitigation topic, visit your repo's landing page and select "manage topics."