IF-Guide

IF-Guide: Desintoxicación de LLMs guiada por funciones de influencia NeurIPS 2025 Este repositorio contiene el código de IF-Guide, la técnica de desintoxicación de LLMs presentada en nuestro artículo: IF-Guide: Influence Function-Guided Detoxification of LLMs Zachary Coalson , Juhan Bae, Nicholas...

IF-GUIDE: Influence Function-Guided Detoxification of LLMs

We study how training data contributes to the emergence of toxic behaviors in large-language models. Most prior work on reducing model toxicity adopts $reactive$ approaches, such as fine-tuning pre-trained and potentially toxic models to align them with human values. In contrast, we propose a...