Non Toxic
Research on "non-toxic" language focuses on detecting and mitigating harmful content generated by large language models (LLMs), particularly toxic, biased, and offensive language. Current efforts concentrate on developing robust detection models using transformer architectures like BERT and LLMs, exploring methods to reduce toxicity during model training and prompting, and creating comprehensive benchmark datasets reflecting diverse languages and cultural contexts. This research is crucial for ensuring the safe and ethical deployment of LLMs in various applications, mitigating the risks of harmful content generation and promoting responsible AI development.
Papers
October 6, 2022
May 15, 2022
May 5, 2022
May 1, 2022
April 30, 2022
April 19, 2022
March 6, 2022
February 26, 2022
February 7, 2022
December 15, 2021
December 7, 2021
November 15, 2021