# DRAGON: Robust Classification for Very Large Collections of Software Repositories

arXiv: https://arxiv.org/abs/2602.09071 · HTML: https://mlsanity.com/paper/2602.09071/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 43 references that are in the map.
- First posted on arXiv 9 Feb 2026.

## Builds on (most cited first, 17 of 17)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [CodeBERT: A Pre-Trained Model for Programming and Natural Languages](https://mlsanity.com/paper/2002.08155/) (2020)
- [Multimodal Machine Learning: A Survey and Taxonomy](https://mlsanity.com/paper/1705.09406/) (2017)
- [Coordinate Descent Algorithms](https://mlsanity.com/paper/1502.04759/) (2015)
- [Class-Conditional Conformal Prediction with Many Classes](https://mlsanity.com/paper/2306.09335/) (2023)
- [Topic Recommendation for Software Repositories using Multi-label Classification Algorithms](https://mlsanity.com/paper/2010.09116/) (2020)
- [HiGitClass: Keyword-Driven Hierarchical Classification of GitHub Repositories](https://mlsanity.com/paper/1910.07115/) (2019)
- [Training Over-parameterized Models with Non-decomposable Objectives](https://mlsanity.com/paper/2107.04641/) (2021)
- [Adapting Neural Text Classification for Improved Software Categorization](https://mlsanity.com/paper/1806.01742/) (2018)
- [Magika: AI-Powered Content-Type Detection](https://mlsanity.com/paper/2409.13768/) (2024)
- [Asymmetric Reinforcing against Multi-modal Representation Bias](https://mlsanity.com/paper/2501.01240/) (2025)
- [GitRanking: A Ranking of GitHub Topics for Software Classification using Active Sampling](https://mlsanity.com/paper/2205.09379/) (2022)
- [Repo2Vec: A Comprehensive Embedding Approach for Determining Repository Similarity](https://mlsanity.com/paper/2107.05112/) (2021)
- [Semantically-enhanced Topic Recommendation System for Software Projects](https://mlsanity.com/paper/2206.00085/) (2022)
- [The penumbra of open source: projects outside of centralized platforms are longer maintained, more academic and more collaborative](https://mlsanity.com/paper/2106.15611/) (2021)
- [LEGION: Harnessing Pre-trained Language Models for GitHub Topic Recommendations with Distribution-Balance Loss](https://mlsanity.com/paper/2403.05873/) (2024)

## Built on by (most cited first, 1 of 1)

- [Industry Classification of GitHub Repositories Using the North American Industry Classification System (NAICS)](https://mlsanity.com/paper/2607.06505/) (2026)

## Abstract (arXiv)

The ability to automatically classify source code repositories with ''topics'' that reflect their content and purpose is very useful, especially when navigating or searching through large software collections. However, existing approaches often rely heavily on README files and other metadata, which are frequently missing, limiting their applicability in real-world large-scale settings. We present DRAGON, a repository classifier designed for very large and diverse software collections. It operates entirely on lightweight signals commonly stored in version control systems: file and directory names, and optionally the README when available. In repository classification at scale, DRAGON improves F1@5 from 54.8% to 60.8%, surpassing the state of the art. DRAGON remains effective even when README files are absent, with performance degrading by only 6% w.r.t. when they are present. This robustness makes it practical for real-world settings where documentation is sparse or inconsistent. Furthermore, many of the remaining classification errors are near misses, where predicted labels are semantically close to the correct topics. This property increases the practical value of the predictions in real-world software collections, where suggesting a few related topics can still guide search and discovery. As a byproduct of developing DRAGON, we also release the largest open dataset to date for repository classification, consisting of 825 thousand repositories with associated ground-truth topics, sourced from the Software Heritage archive, providing a foundation for future large-scale and language-agnostic research on software repository understanding.