Chain-of-Table is a reasoning chain framework for table understanding, specifically designed for tasks such as question answering based on tables and fact verification. It uses tabular data as part of the reasoning chain and guides large language models to perform operation generation and table updates in a contextual learning manner, forming a continuous reasoning chain that demonstrates the reasoning process for given table questions. This reasoning chain contains structured information about intermediate results, enabling more accurate and reliable predictions. Chain-of-Table has achieved new state-of-the-art performance on multiple benchmark tests, including WikiTQ, FeTaQA, and TabFact.