BoilerBot: A Reliable Task-Oriented Chatbot Enhanced with Large Language Models

Abstract

This paper outlines the design and deployment of BoilerBot: a task-oriented multi-modal conversational agent developed for the Alexa Prize TaskBot 2 competition. BoilerBot features flexible response generation, leveraging Large Language Models (LLMs) to enable adaptable user experiences. We discuss our novel contributions towards advancing state-of-the-art task-oriented conversational agents, highlighting user-facing challenges and proposing fault-tolerant, iterative solutions for carefully guided workflows that enable Alexa users to maximize BoilerBot’s functionality.

Publication
2nd Proceedings of Alexa Prize TaskBot (Alexa Prize 2023)
Jinen Setpal
Jinen Setpal
ML Research Intern @ Apple

Deep Learning Optimization Theory & Intrinsic Interpretability