1 code implementation • 22 Aug 2023 • Noel Ngu, Nathaniel Lee, Paulo Shakarian
In this paper, we present measures for quantification of error in the response of a large language model based on the diversity of responses to a given prompt - hence independent of the underlying application.
1 code implementation • 23 Feb 2023 • Paulo Shakarian, Abhinav Koyyalamudi, Noel Ngu, Lakshmivihari Mareedu
We study the performance of a commercially available large language model (LLM) known as ChatGPT on math word problems (MWPs) from the dataset DRAW-1K.