Abstract
“Rapid advancements in artificial intelligence have led to the widespread deploy ment of language models across various domains, from customer service to content generation, where the generation of accurate and ethically sound responses is p aramount. Addressing the dual challenges of toxic content generation and the need for sophisticated reasoning, this research introduces a novel approach that in tegrates targeted pruning techniques to enhance multi-hop reasoning while simult aneously reducing the presence of toxic knowledge. By fine-tuning the pre-trained Llama model, the study explores how selective pruning of toxic pathways can im prove the model’s accuracy, response diversity, and robustness against adversarial prompts.
NETL
NSTL