Metraipolis › archive › dc_88b239ab5baf

Elite-Weighted Supervised Fine-tuning for Goal-Directed Molecular Optimization — arXiv:2609.00189

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.00189
License (as archived)
CC0-1.0
Archived
2026-10-17T06:52:02.046Z
SHA-256 of the archived text
1f8aabd1c3e9582f482e30601d0bcf1b4ff6a8fa66e7cf525f107575a8d55a82
Stable URL of this page
https://metraipolis.com/documento/dc_88b239ab5baf

How to cite

Elite-Weighted Supervised Fine-tuning for Goal-Directed Molecular Optimization — arXiv:2609.00189 — arXiv (en), https://arxiv.org/abs/2609.00189 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_88b239ab5baf (Metraipolis archive dc_88b239ab5baf, sha256 1f8aabd1c3e9582f482e30601d0bcf1b4ff6a8fa66e7cf525f107575a8d55a82)
BibTeX
@misc{dc88b239ab5baf,
  title = {Elite-Weighted Supervised Fine-tuning for Goal-Directed Molecular Optimization — arXiv:2609.00189},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.00189},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_88b239ab5baf; sha256 1f8aabd1c3e9582f482e30601d0bcf1b4ff6a8fa66e7cf525f107575a8d55a82},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00189. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Goal-directed optimization is essential for steering molecular generators to propose candidates with desired properties. However, it is often implemented with policy-gradient reinforcement learning, which requires a generation-trajectory log-probability whose form depends on the model architecture and generation procedure. This makes an optimizer difficult to reuse across architectures and conditional generative designs. Supervised fine-tuning needs none of that machinery, but its update is driven by a fixed dataset, so the reward never enters the update. We introduce Elite-Weighted Supervised Fine-tuning (EW-SFT), which uses reward to guide elite selection of high-scoring molecules, and updates the model by its own pretraining loss on that set. Ablations show that reward information is passed primarily through elite selection, rather than through continuous weighting within the selected set. Because the update consumes only scored molecules and the model's native loss, the same rule applies across autoregressive, masked-diffusion, and discrete-flow generators, and across de novo, motif-extension, and linker-design tasks. Under a fixed budget of 3D shape alignment oracle calls on two kinase reference compounds, EW-SFT consistently outperforms the corresponding native optimizers. It further improves goal-directed optimization under a 2D similarity oracle on four held-out references and achieves comparable performance on a sample-efficiency benchmark without a trajectory-level RL formulation. These results demonstrate that EW-SFT is a unified and effective optimizer across molecular generators, design constraints, references, and oracles.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.00189 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T06:41:55.771Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).