Sample Efficient Generative Molecular Optimization with Joint Self-Improvement
Serra Korkmaz, Adam Izdebski, Jonathan Pirnay, Rasmus Møller-Larsen, Michal Kmicikiewicz, Pankhil Gawade, Dominik G. Grimm, Ewa Szczurek · 2026
Read Sample Efficient Generative Molecular Optimization with Joint Self-Improvement by Serra Korkmaz, Adam Izdebski, Jonathan Pirnay, Rasmus Møller-Larsen, Michal Kmicikiewicz, Pankhil Gawade, Dominik G. Grimm, Ewa Szczurek (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/arxiv-2602-10984.txt.
Description
Designing new molecules is slow because good candidates are rare and costly to test. Serra Korkmaz and seven co-authors propose Joint Self-Improvement, a method that pairs a molecule generator with a property predictor in one joint model. The pairing keeps the predictor honest as search drifts to new regions.
A self-improving sampling step then steers the generator with the predictor at run time, so fresh candidates lean toward strong predicted scores. Tests on offline and online molecule tasks show the method beats leading rivals when test budgets are tight. The gain matters most where each lab test costs time and money, and where a few strong candidates beat many weak ones.
Catalog record
| CREATOR | Serra Korkmaz, Adam Izdebski, Jonathan Pirnay, Rasmus Møller-Larsen, Michal Kmicikiewicz, Pankhil Gawade, Dominik G. Grimm, Ewa Szczurek |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-10984