Towards Zero-Shot and Few-Shot Table Question Answering using GPT-3
This work addresses table question answering for natural language processing applications, but it is incremental as it applies an existing method to a new dataset with basic enhancements.
The paper tackled table question answering by using GPT-3 with zero-shot and few-shot learning, finding that it could answer lookup and comparison queries from serialized tables without fine-tuning, with accuracy improved by prompt engineering and intermixing text.
We present very early results on using GPT-3 to perform question answering on tabular data. We find that stock pre-trained GPT-3 is able to zero-shot learn the table structure from a serialized JSON array-of-arrays representation, and able to answer lookup queries and simple comparison questions in natural language without any fine-tuning. We further find that simple prompt engineering to include few-shot static Q&A examples significantly improves accuracy. Lastly, we find that intermixing passage text improves accuracy even further on heterogeneous data. We apply our approach on a novel dataset of simple tables in newspaper infographics with promising results. Overall, we find much cause for optimism in this basic approach.