Task-Oriented Dialogue with In-Context Learning
This addresses the problem of scaling and simplifying chatbot development for industry practitioners, though it appears incremental as it builds on existing LLM capabilities.
The paper tackles the challenge of building task-oriented dialogue systems by combining large language models (LLMs) with deterministic business logic, using a domain-specific language (DSL) for translation. The result shows that developing chatbots with this system requires significantly less effort than intent-based NLU approaches and can handle complex dialogues more successfully.
We describe a system for building task-oriented dialogue systems combining the in-context learning abilities of large language models (LLMs) with the deterministic execution of business logic. LLMs are used to translate between the surface form of the conversation and a domain-specific language (DSL) which is used to progress the business logic. We compare our approach to the intent-based NLU approach predominantly used in industry today. Our experiments show that developing chatbots with our system requires significantly less effort than established approaches, that these chatbots can successfully navigate complex dialogues which are extremely challenging for NLU-based systems, and that our system has desirable properties for scaling task-oriented dialogue systems to a large number of tasks. We make our implementation available for use and further study.