Measuring AI Progress in Drug Discovery: A Reproducible Leaderboard for the Tox21 Challenge
Antonia Ebner, Christoph Bartmann, Sonja Topf, Günter Klambauer, Johannes Schimunek, Sohvi Luukkonen · 2026
Read Measuring AI Progress in Drug Discovery: A Reproducible Leaderboard for the Tox21 Challenge by Antonia Ebner, Christoph Bartmann, Sonja Topf, Günter Klambauer, Johannes Schimunek, Sohvi Luukkonen (2026) 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/pmc-13625945.txt.
Description
Antonia Ebner and five colleagues published a 2026 paper on measuring progress in AI for drug discovery, using the Tox21 challenge as their test case. Tox21 asks computer models to predict whether chemicals are toxic. In 2015, deep learning models beat older methods in this challenge, a turning point often compared with the ImageNet moment in computer vision.
After the challenge ended, its data set was folded into wider benchmark collections, and scores became hard to compare across years. The authors built a reproducible leaderboard so new methods can be judged on the same footing as old ones. A stable leaderboard shows whether AI in drug discovery is truly improving or merely changing its test conditions.
Catalog record
| CREATOR | Antonia Ebner, Christoph Bartmann, Sonja Topf, Günter Klambauer, Johannes Schimunek, Sohvi Luukkonen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |