Semantic Parsing in Limited Resource Conditions

TL;DR

This thesis presents innovative methods for semantic parsing under limited data and computational resources, including synthetic data generation, knowledge transfer, active learning, and continual learning, to enhance performance across diverse scenarios.

Contribution

It introduces novel techniques for data augmentation, domain adaptation, multilingual parsing, and resource-efficient training, addressing key challenges in resource-constrained semantic parsing.

Findings

Synthetic training data improves parsing accuracy in zero-resource scenarios.

Knowledge transfer enhances target domain parsing with limited data.

Continual learning maintains performance while reducing training time.

Abstract

This thesis explores challenges in semantic parsing, specifically focusing on scenarios with limited data and computational resources. It offers solutions using techniques like automatic data curation, knowledge transfer, active learning, and continual learning. For tasks with no parallel training data, the thesis proposes generating synthetic training examples from structured database schemas. When there is abundant data in a source domain but limited parallel data in a target domain, knowledge from the source is leveraged to improve parsing in the target domain. For multilingual situations with limited data in the target languages, the thesis introduces a method to adapt parsers using a limited human translation budget. Active learning is applied to select source-language samples for manual translation, maximizing parser performance in the target language. In addition, an alternative method is also proposed to utilize machine translation services, supplemented by human-translated data, to train a more effective parser. When computational resources are limited, a continual learning approach is introduced to minimize training time and computational memory. This maintains the parser's efficiency in previously learned tasks while adapting it to new tasks, mitigating the problem of catastrophic forgetting. Overall, the thesis provides a comprehensive set of methods to improve semantic parsing in resource-constrained conditions.