Improving Conversational Recommendation Systems via Bias Analysis and Language-Model-Enhanced Data Augmentation
Submission Type: Regular Long Paper
Submission Track: Dialogue and Interactive Systems
Submission Track 2: Natural Language Generation
Keywords: Conversational Recommendation, Bias Mitigation, Generative Data, Data Augmentation
TL;DR: The use of synthetic recommendation dialogues with movie variants to address various biases within conversational recommendation as well as the introduction of additional evaluation metrics, joint with rich analysis.
Abstract: Conversational Recommendation System (CRS) is a rapidly growing research area that has gained significant attention alongside advancements in language modelling techniques. However, the current state of conversational recommendation faces numerous challenges due to its relative novelty and limited existing contributions. In this study, we delve into benchmark datasets for developing CRS models and address potential biases arising from the feedback loop inherent in multi-turn interactions, including selection bias and multiple popularity bias variants. Drawing inspiration from the success of generative data via using language models and data augmentation techniques, we present two novel strategies, 'Once-Aug' and 'PopNudge', to enhance model performance while mitigating biases. Through extensive experiments on ReDial and TG-ReDial benchmark datasets, we show a consistent improvement of CRS techniques with our data augmentation approaches and offer additional insights on addressing multiple newly formulated biases.
Submission Number: 1044
Loading