Position: More Rigorous Software Engineering Would Improve Reproducibility in Machine Learning Research
Moritz Wolter, Lokesh Veeramacheneni · 2025
Read Position: More Rigorous Software Engineering Would Improve Reproducibility in Machine Learning Research by Moritz Wolter, Lokesh Veeramacheneni (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-00902.txt.
Description
Moritz Wolter and co-authors argue that machine learning research needs stricter software habits, in a 2025 position paper. Reproducing a result is basic science, yet code behind headline papers often lacks the basics: pinned versions, tests, and clear structure. The authors set out to measure the gap.
They surveyed code repositories tied to papers at top venues over the past decade. The survey finds wide neglect of standard practices, alongside pockets of progress. Poor habits waste other groups' time and erode trust in published claims. The paper closes with concrete steps for authors, reviewers, and venues to raise the floor. Its plea is that software craft is not housekeeping but part of the science itself.
Catalog record
| CREATOR | Moritz Wolter, Lokesh Veeramacheneni |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-00902