Ml4Se: Curated Research Resources for ML in Software Engineering

This curated repository provides a collection of research papers, theses, datasets, and tools focused on Machine Learning for Software Engineering (ML4SE). Maintained by the Delft University of Technology (saltudelft) and updated as recently as November 2025, it serves as a bibliographic and resource index for the AI4Code community. The collection spans multiple categories including deep learning for code and LLM-based software engineering.

Use Cases

• Conducting literature reviews for ML4SE by filtering the 'papers' and 'theses' categories
• Identifying benchmarking tools by comparing entries in the 'tools' section
• Locating training data for code models by exploring the 'datasets' list

Strengths

• Maintained by the Software Engineering Research Group at TU Delft
• Includes specialized resource types such as theses and software tools often omitted from standard bibliographies
• Updated as of November 2025

Limitations

• Lack of structured metadata fields for automated filtering
• Subjective selection criteria inherent to curated lists

Provenance

Source

Delft University of Technology (saltudelft)

Collection Method

Curated

Freshness

Highly current, with the last update recorded in November 2025.

Geography

Global