One vs. Many: Comprehending Accurate Information from Multiple Erroneous and Inconsistent AI Generations

arXiv:2405.05581v1 Announce Type: cross
Abstract: As Large Language Models (LLMs) are nondeterministic, the same input can generate different outputs, some of which may be incorrect or hallucinated. If run again, the LLM may correct itself and produce the correct answer. Unfortunately, most LLM-powered systems resort to single results which, correct or not, users accept. Having the LLM produce multiple outputs may help identify disagreements or alternatives. However, it is not obvious how the user will interpret conflicts or inconsistencies. To …

abstract arxiv cs.ai cs.cl cs.hc generate information language language models large language large language models llm llms multiple nondeterministic systems type