Rewarding Chatbots for Real-World Engagement with Millions of Users

TL;DR

This paper presents a reward-based training method for social chatbots that significantly improves user engagement and retention by using human feedback and automatic pseudo-labels, validated through large-scale A/B testing.

Contribution

It introduces a novel reward modeling approach using pseudo-labels from user interactions to enhance chatbot engagement and retention.

Findings

Mean conversation length increased by up to 70%.

User retention improved by over 30%.

Effective use of human feedback for training chatbots.

Abstract

The emergence of pretrained large language models has led to the deployment of a range of social chatbots for chitchat. Although these chatbots demonstrate language ability and fluency, they are not guaranteed to be engaging and can struggle to retain users. This work investigates the development of social chatbots that prioritize user engagement to enhance retention, specifically examining the use of human feedback to efficiently develop highly engaging chatbots. The proposed approach uses automatic pseudo-labels collected from user interactions to train a reward model that can be used to reject low-scoring sample responses generated by the chatbot model at inference time. Intuitive evaluation metrics, such as mean conversation length (MCL), are introduced as proxies to measure the level of engagement of deployed chatbots. A/B testing on groups of 10,000 new daily chatbot users on the Chai Research platform shows that this approach increases the MCL by up to 70%, which translates to a more than 30% increase in user retention for a GPT-J 6B model. Future work aims to use the reward model to realise a data fly-wheel, where the latest user conversations can be used to alternately fine-tune the language model and the reward model.