UnBias-Plus: Detect, Explain, and Rewrite Bias
The paper presents UnBias-Plus, an open-source toolkit for detecting and addressing bias in natural language content.
UnBias-Plus is novel in its comprehensive approach to bias detection, providing a unified toolkit for classification, localization, rewriting, and explanation.
Before reading this…
Applications
- →journalism
- →education
- →AI research
To understand this paper, make sure you know these concepts first:
- natural language processingfind papers →
- machine learningfind papers →
Abstract
More Like ThisBias 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.