Handling Ontology Gaps in Semantic Parsing

TL;DR

This paper introduces the Hallucination Simulation Framework to analyze and detect hallucinations in neural semantic parsing models, especially addressing ontology gaps, thereby improving model reliability in question answering systems.

Contribution

It proposes a novel framework for simulating NSP hallucinations and a new detection strategy that leverages the model's computational graph to identify ontology gaps and errors.

Findings

Improved hallucination detection F1-Score by ~21%.

Enhanced recognition of ontology gaps and out-of-domain utterances.

First approach addressing ontology gaps in closed-ontology NSP.

Abstract

The majority of Neural Semantic Parsing (NSP) models are developed with the assumption that there are no concepts outside the ones such models can represent with their target symbols (closed-world assumption). This assumption leads to generate hallucinated outputs rather than admitting their lack of knowledge. Hallucinations can lead to wrong or potentially offensive responses to users. Hence, a mechanism to prevent this behavior is crucial to build trusted NSP-based Question Answering agents. To that end, we propose the Hallucination Simulation Framework (HSF), a general setting for stimulating and analyzing NSP model hallucinations. The framework can be applied to any NSP task with a closed-ontology. Using the proposed framework and KQA Pro as the benchmark dataset, we assess state-of-the-art techniques for hallucination detection. We then present a novel hallucination detection strategy that exploits the computational graph of the NSP model to detect the NSP hallucinations in the presence of ontology gaps, out-of-domain utterances, and to recognize NSP errors, improving the F1-Score respectively by ~21, ~24% and ~1%. This is the first work in closed-ontology NSP that addresses the problem of recognizing ontology gaps. We release our code and checkpoints at https://github.com/amazon-science/handling-ontology-gaps-in-semantic-parsing.