Few-Shot Table-to-Text Generation with Prototype Memory
This addresses the data-hungry limitation of neural models for table-to-text generation, making it more applicable in real-world settings where large training data is scarce.
The paper tackles the problem of table-to-text generation in few-shot scenarios by proposing a Prototype-to-Generate framework, which improves model performance across three benchmark datasets with significant gains in evaluation metrics.
Neural table-to-text generation models have achieved remarkable progress on an array of tasks. However, due to the data-hungry nature of neural models, their performances strongly rely on large-scale training examples, limiting their applicability in real-world applications. To address this, we propose a new framework: Prototype-to-Generate (P2G), for table-to-text generation under the few-shot scenario. The proposed framework utilizes the retrieved prototypes, which are jointly selected by an IR system and a novel prototype selector to help the model bridging the structural gap between tables and texts. Experimental results on three benchmark datasets with three state-of-the-art models demonstrate that the proposed framework significantly improves the model performance across various evaluation metrics.