ahmedallam/BiasDPO

Citation text from the card

@inproceedings{allam-2024-biasdpo,
    title = "{B}ias{DPO}: Mitigating Bias in Language Models through Direct Preference Optimization",
    author = "Allam, Ahmed",
    editor = "Fu, Xiyan  and
      Fleisig, Eve",
    booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)",
    month = aug,
    year = "2024",
    address = "Bangkok, Thailand",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.acl-srw.7",
    pages = "71--79",
    abstract = "Large Language Models (LLMs) have become pivotal in advancing natural language processing, yet their potential to perpetuate biases poses significant concerns. This paper introduces a new framework employing Direct Preference Optimization (DPO) to mitigate gender, racial, and religious biases in LLM-generated English text. By developing a loss function that favors less biased over biased completions, our approach cultivates a preference for respectful and non-discriminatory language in LLMs. We also contribute a manually designed dataset for training LLMs to recognize and correct biases. This dataset encompasses a diverse range of prompts paired with both biased and unbiased completions. Implementing this approach on the Microsoft Phi-2 model, we demonstrate substantial reductions in biased outputs as our model outperforms the baseline model on almost all bias benchmarks. Our model also achieves better 

All datasets marked Permissive or share-alike | JSON