Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding

TL;DR

Neural Path Hunter enhances dialogue system faithfulness by refining generated responses with knowledge graph paths, significantly reducing hallucinations without retraining the base model.

Contribution

It introduces a generate-then-refine approach using a fact critic and chain of neural language models to improve factual accuracy in dialogue responses.

Findings

20.35% relative improvement in faithfulness over baseline

Effective application without retraining the dialogue model

Validated on OpenDialKG dataset with multiple metrics

Abstract

Dialogue systems powered by large pre-trained language models (LM) exhibit an innate ability to deliver fluent and natural-looking responses. Despite their impressive generation performance, these models can often generate factually incorrect statements impeding their widespread adoption. In this paper, we focus on the task of improving the faithfulness -- and thus reduce hallucination -- of Neural Dialogue Systems to known facts supplied by a Knowledge Graph (KG). We propose Neural Path Hunter which follows a generate-then-refine strategy whereby a generated response is amended using the k-hop subgraph of a KG. Neural Path Hunter leverages a separate token-level fact critic to identify plausible sources of hallucination followed by a refinement stage consisting of a chain of two neural LM's that retrieves correct entities by crafting a query signal that is propagated over the k-hop subgraph. Our proposed model can easily be applied to any dialogue generated responses without retraining the model. We empirically validate our proposed approach on the OpenDialKG dataset against a suite of metrics and report a relative improvement of faithfulness over dialogue responses by 20.35% based on FeQA (Durmus et al., 2020).