This repository provides a comprehensive survey of Neural Code Intelligence, acting as a curated reading list and resource hub for researchers and practitioners in the field. It aims to consolidate paradigms, advances, and future directions in using neural networks for code-related tasks, offering a valuable starting point for understanding the landscape.

How It Works

The project is structured around a survey paper, "A Survey of Neural Code Intelligence: Paradigms, Advances and Beyond," which serves as the central organizing principle. It categorizes and lists relevant research papers, development timelines, and emerging trends in areas like code generation, repair, reasoning, and evaluation, facilitating a structured approach to learning and research.

Quick Start & Requirements

• No installation or execution commands are provided as this is a curated list of resources, not a software package.
• Requirements are conceptual: an interest in AI and code intelligence research.
• Links to the survey paper and related slides are available within the README.

Highlighted Details

• Regularly updated with new research papers and topics, including code preference learning and LLM evaluations.
• Features a visual timeline of code intelligence development and recent work.
• Organizes resources into categories such as Language Models for Code, Evaluations, Code Repair, and Code-Based NLP.
• Includes links to slides, talks, blogs, and a citation format for the survey paper.

Maintenance & Community

The project is an open collaborative research effort with regular updates. It welcomes pull requests for missing papers. A star history is provided, indicating community interest.

Licensing & Compatibility

The repository itself does not appear to have a specific license mentioned for its curated content. The underlying research papers referenced will have their own respective licenses.

Limitations & Caveats

This repository is a curated list of research papers and resources, not a functional software tool. Its utility is dependent on the completeness and accuracy of the survey paper and the community's contributions.