Work

Less is More: Mitigate Spurious Correlations for Open-Domain Dialogue Response Generation Models by Causal Discovery

Year: 2023

Type: article

Abstract: In this paper, we conduct the first study on spurious correlations for open-domain response generation models based on a corpus CGDialog curated by ourselves. The current models indeed suffer from spurious correlations and have a tendency to generate irrelevant and generic responses. Inspired by causal discovery algorithms, we propose a novel model-agnostic method for training and inference using a conditional independence classifier. The classifier is trained by a constrained self-training method, coined ConSTrain, to overcome data sparsity. The experimental results based on both human and automatic evaluation show that our method significantly outperforms the competitive baselines in terms of relevance, informativeness, and fluency.

Source: Transactions of the Association for Computational Linguistics

DOI: 10.1162/tacl_a_00561

Authors: Tao Feng, Lizhen Qu, Gholamreza Haffari

Institution: Monash University

Language: English

Topic: Speech and dialogue systems

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: diamond

Funder: Defense Advanced Research Projects Agency

Awards: G2859168290, G3717735613, G4713059963