This repository is a curated list of academic papers focusing on fairness, accountability, transparency, and ethics in Natural Language Processing (NLP) and multi-modal models. It serves as a comprehensive reference for researchers and practitioners interested in understanding, detecting, and mitigating biases within AI systems.

How It Works

The project organizes papers into thematic categories such as Surveys, Social Impact of Biases, Data/Models/Metrics, Bias Amplification, Detection, Mitigation, and specific NLP tasks like Generation and Machine Translation. It also includes sections for multi-modal settings, tutorials, and relevant conferences, providing a structured overview of the research landscape.

Quick Start & Requirements

This is a curated list of papers, not a software package. No installation or execution is required.

Highlighted Details

• Extensive categorization of papers covering a wide range of fairness topics in NLP.
• Includes seminal works and recent research from top NLP venues.
• Points to tutorials, datasets, and workshops for deeper engagement.
• Covers both traditional NLP tasks and emerging multi-modal applications.

Maintenance & Community

The repository is maintained by Christina Chance, Yixin Wan, Jieyu Zhao, Emily Sheng, Sunipa Dev, Yu (Hope) Hou, Nanyun (Violet) Peng, and Kai-Wei Chang. Contributions are welcomed via pull requests or email.

Licensing & Compatibility

The content is presented for informational and research purposes. Specific paper licenses would apply to the individual works listed.

Limitations & Caveats

The authors acknowledge that the list may not be exhaustive and encourage community contributions to improve its completeness.