SWE-Milestone
- Type
- dataset
- Venue
- DeepCommit-ai
- Year
- 2026
- Source
- huggingface
- Access
- free
- Added
- 2026-07-17T20:01:18.204153+00:00
- Verified
- 2026-07-17T20:01:18.204153+00:00
Summary
Software evolution itineraries (as Milestone DAGs) extracted from real-world repositories for AI agent evaluation. Used by [SWE-Milestone](https://github.com/DeepCommit-ai/SWE-Milestone). [[Paper]](https://arxiv.org/abs/2603.13428)
Keywords
hf-dataset language-modeling · -code code benchmark agents software-engineering evaluation has-paper
Topics
Language Modeling, Code
Research notes
- downloads=1034; likes=4