Memory-Based Semantic Parsing

TL;DR

This paper introduces a memory-based model for context-dependent semantic parsing that uses an external memory to manage sequential utterance meanings, improving performance across benchmarks without task-specific decoders.

Contribution

It proposes a novel external memory approach with a context memory controller for semantic parsing, differing from prior methods that rely on copying or modifying previous parses.

Findings

Improved accuracy on three semantic parsing benchmarks.

Effective handling of context-dependent information.

No need for task-specific decoders.

Abstract

We present a memory-based model for context-dependent semantic parsing. Previous approaches focus on enabling the decoder to copy or modify the parse from the previous utterance, assuming there is a dependency between the current and previous parses. In this work, we propose to represent contextual information using an external memory. We learn a context memory controller that manages the memory by maintaining the cumulative meaning of sequential user utterances. We evaluate our approach on three semantic parsing benchmarks. Experimental results show that our model can better process context-dependent information and demonstrates improved performance without using task-specific decoders.