Conversational programming refers to a programming approach that aims to make human-computer interactions more natural and intuitive, often by leveraging natural language processing (NLP) techniques. The goal is to allow users to interact with computers in a conversational manner, potentially making software more accessible to those without technical expertise.
Here is a alexa-style conversational programming approach, where the user is asked to provide a natural language query, and the computer responds with an appropriate response presented at re-invent
Top research papers in this area often come from conferences like ACL, EMNLP, and NeurIPS, among others. Here are some influential papers:
- “Dialog-to-Action: Converting Conversational Dialogues to Executable Programs” - Various authors from Microsoft Research and University of Michigan
- “Conversational Program Synthesis” - Various authors, often affiliated with institutions like Stanford and MIT
- “Building Conversational Agents with Transformer Neural Networks” - Research from Google Brain
- “Towards End-to-End Learning for Dialog State Tracking and Management using Deep Reinforcement Learning” - Research from Facebook AI Research (FAIR)
As we build LIRA, we are interested in exploring this area of conversational programming. We are also interested in exploring the use of Large Language Models (LMs) for conversational programming.