UnBias-Plus: Detect, Explain, and Rewrite Bias
2026-06-22 • Computation and Language
Computation and LanguageArtificial IntelligenceSoftware Engineering
AI summaryⓘ
The authors created UnBias-Plus, a free tool to help find and fix bias in text, including both human and AI writing. Their tool can detect different types of bias in parts of the text, highlight exactly where the bias is, explain its decisions, and rewrite biased sentences to be neutral. It is easy to use through various methods like Python and web interfaces. They also shared all their code, models, and data for others to use and improve.
bias detectionnatural language processingmulti-class classificationspan localizationneutral rewritingexplainable AIopen-source toolkitPython APIREST APItext analysis
Authors
Ahmed Y. Radwan, Ahmed ElKady, Sindhuja Chaduvula, Mohamed Hafez, Amrit Krishnan, Shaina Raza
Abstract
Bias in natural language remains a persistent challenge in both human-written and AI-generated content, affecting domains such as journalism, education, and AI research. Most existing detection methods identify only the presence of bias, with limited support for granular detection, interpretable explanations, neutral rewriting, and openly available trained models. We present UnBias-Plus, an open-source toolkit unifying (1) segment-level multi-class bias classification, (2) biased span localization, (3) neutral text rewriting, and (4) reasoning for each decision. Available via Python, CLI, REST API, and web interfaces, UnBias-Plus supports accessible bias analysis. The toolkit, source code, models, datasets, and documentation are publicly available.