Learning Improvised Chatbots from Adversarial Modifications of Natural Language Feedback

TL;DR

This paper introduces a generative adversarial approach to refine user feedback in chatbots, transforming noisy feedback into plausible responses, thereby enhancing chatbot performance significantly on the Personachat dataset.

Contribution

The paper presents a novel adversarial model that converts noisy natural language feedback into effective training responses, improving chatbot performance.

Findings

Chatbot performance improved from 69.94% to 75.96% in response ranking.

The adversarial feedback augmentation significantly enhances chatbot training.

The method effectively handles extraneous sequences in user feedback.

Abstract

The ubiquitous nature of chatbots and their interaction with users generate an enormous amount of data. Can we improve chatbots using this data? A self-feeding chatbot improves itself by asking natural language feedback when a user is dissatisfied with its response and uses this feedback as an additional training sample. However, user feedback in most cases contains extraneous sequences hindering their usefulness as a training sample. In this work, we propose a generative adversarial model that converts noisy feedback into a plausible natural response in a conversation. The generator's goal is to convert the feedback into a response that answers the user's previous utterance and to fool the discriminator which distinguishes feedback from natural responses. We show that augmenting original training data with these modified feedback responses improves the original chatbot performance from 69.94% to 75.96% in ranking correct responses on the Personachat dataset, a large improvement given that the original model is already trained on 131k samples.