Programming by Example and Text-to-Code Translation for Conversational Code Generation
This addresses the issue of rigid dialogue paths in conversational AI, offering a more flexible approach for developers and users, though it appears incremental as it combines existing techniques.
The paper tackles the problem of deterministic dialogue systems by proposing MPaTHS, a method that integrates Programming by Example and text-to-code translation to synthesize general programs from natural language, and demonstrates its application to task-oriented dialogue.
Dialogue systems is an increasingly popular task of natural language processing. However, the dialogue paths tend to be deterministic, restricted to the system rails, regardless of the given request or input text. Recent advances in program synthesis have led to systems which can synthesize programs from very general search spaces, e.g. Programming by Example, and to systems with very accessible interfaces for writing programs, e.g. text-to-code translation, but have not achieved both of these qualities in the same system. We propose Modular Programs for Text-guided Hierarchical Synthesis (MPaTHS), a method for integrating Programming by Example and text-to-code systems which offers an accessible natural language interface for synthesizing general programs. We present a program representation that allows our method to be applied to the problem of task-oriented dialogue. Finally, we demo MPaTHS using our program representation.