deep learning has become the dominant approach in coping with various tasks in Natural LanguageProcessing (NLP). Although text inputs are typically represented as a sequence of tokens, there isa rich variety of NLP problems that can be best expressed with a graph structure. As a result
NLGraph is a benchmark for large language models to solve graph-based problems in natural language, where two instruction-based approaches - Build-a-Graph Prompting and Algorithmic Prompting - are proposed, improving the LLMs' performance by 3.07% to 16.85% across multiple tasks and settings.