Evaluating Large Language Models in Semantic Parsing for Conversational Question Answering over Knowledge Graphs

TL;DR

This paper assesses the ability of large language models to generate graph queries for knowledge-based conversational question answering, highlighting the impact of prompting and fine-tuning on performance.

Contribution

It provides a comprehensive evaluation of large language models on semantic parsing for conversational QA over knowledge graphs, including analysis of prompting techniques and model size effects.

Findings

Large language models can generate graph queries from dialogues.

Few-shot prompting and fine-tuning improve model performance.

Smaller models benefit significantly from tuning techniques.

Abstract

Conversational question answering systems often rely on semantic parsing to enable interactive information retrieval, which involves the generation of structured database queries from a natural language input. For information-seeking conversations about facts stored within a knowledge graph, dialogue utterances are transformed into graph queries in a process that is called knowledge-based conversational question answering. This paper evaluates the performance of large language models that have not been explicitly pre-trained on this task. Through a series of experiments on an extensive benchmark dataset, we compare models of varying sizes with different prompting techniques and identify common issue types in the generated output. Our results demonstrate that large language models are capable of generating graph queries from dialogues, with significant improvements achievable through few-shot prompting and fine-tuning techniques, especially for smaller models that exhibit lower zero-shot performance.