Axolotl: Fairness through Assisted Prompt Rewriting of Large Language Model Outputs

Pre-trained Large Language Models (LLMs) have significantly advanced natural language processing capabilities but are susceptible to biases present in their training data, leading to unfair outcomes in various applications. While numerous strategies have been proposed to mitigate bias, they often require extensive computational resources and may compromise model performance.In this work, inspired by the concept of query rewriting, we propose prompt rewriting to minimize the unfairness of LLM outputs. We introduce Axolotl, a novel post-processing framework that operates agnostically across tasks and models, leveraging public APIs to interact with LLMs without direct access to internal parameters. Through a three-step process, Axolotl identifies biases, proposes resolutions, and guides the model to self-debias its outputs. This approach minimizes computational costs and preserves model performance, making Axolotl a promising tool for debiasing LLM outputs with broad applicability and ease of use.