Natural Language Outlines for Code: Literate Programming in the LLM Era

We propose using natural language outlines as a novel modality and interaction surface for providing AI assistance to developers throughout the software development process. An NL outline for a code function comprises multiple statements written in concise prose, which partition the code and summari...

Full description

Saved in:

Bibliographic Details
Main Authors:	Shi, Kensen, Altınbüken, Deniz, Anand, Saswat, Christodorescu, Mihai, Grünwedel, Katja, Koenings, Alexa, Naidu, Sai, Pathak, Anurag, Rasi, Marc, Ribeiro, Fredde, Ruffin, Brandon, Sanyam, Siddhant, Tabachnyk, Maxim, Toth, Sara, Tu, Roy, Welp, Tobias, Yin, Pengcheng, Zaheer, Manzil, Chandra, Satish, Sutton, Charles
Format:	Journal Article
Language:	English
Published:	08-08-2024
Subjects:	Computer Science - Artificial Intelligence Computer Science - Computation and Language Computer Science - Human-Computer Interaction Computer Science - Learning Computer Science - Software Engineering
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	We propose using natural language outlines as a novel modality and interaction surface for providing AI assistance to developers throughout the software development process. An NL outline for a code function comprises multiple statements written in concise prose, which partition the code and summarize its main ideas in the style of literate programming. Crucially, we find that modern LLMs can generate accurate and high-quality NL outlines in practice. Moreover, NL outlines enable a bidirectional sync between code and NL, allowing changes in one to be automatically reflected in the other. We discuss many use cases for NL outlines: they can accelerate understanding and navigation of code and diffs, simplify code maintenance, augment code search, steer code generation, and more. We then propose and compare multiple LLM prompting techniques for generating outlines and ask professional developers to judge outline quality. Finally, we present two case studies applying NL outlines toward code review and the difficult task of malware detection.
DOI:	10.48550/arxiv.2408.04820