Reproduce “Roll Up Your Sleeves: Working with a Collaborative and Engaging Task-Oriented Dialogue System” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Lingbo Mo, Shijie Chen, Ziru Chen, Xiang Deng, Ashley Lewis, Sunit Singh, Samuel Stevens, Chang-You Tai and 5 more
TACOBOT introduces a dialogue system designed to guide users through complex, multi-step real-world tasks like cooking recipes while maintaining an engaging, conversational interaction style. This system helps users perform tasks themselves rather than simply retrieving information, with its predecessor placing third in the Alexa Prize TaskBot Challenge.
Key insights
- Guiding users through complex, multi-step tasks collaboratively requires a shift from traditional dialogue management and response generation strategies.
- Large language models can effectively augment limited in-domain data, improving natural language understanding robustness.
- Integrating casual conversation (chit-chat) with task assistance enhances user engagement and naturalness in dialogue systems.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.