Resilience of Large Language Models for Noisy Instructions

arXiv:2404.09754v1 Announce Type: new
Abstract: As the rapidly advancing domain of natural language processing (NLP), large language models (LLMs) have emerged as powerful tools for interpreting human commands and generating text across various tasks. Nonetheless, the resilience of LLMs to handle text containing inherent errors, stemming from human interactions and collaborative systems, has not been thoroughly explored. Our study investigates the resilience of LLMs against five common types of disruptions including 1) ASR (Automatic Speech Recognition) errors, 2) OCR (Optical …

abstract arxiv collaborative cs.cl domain errors human human interactions interactions language language models language processing large language large language models llms natural natural language natural language processing nlp processing resilience stemming systems tasks text tools type