Metraipolis › archive › dc_ccc82f18d892
Fixed-Dimensional Latent Flow for Generating Variable-Size 3D Molecules — arXiv:2609.08333
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.08333
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T13:59:30.790Z
- SHA-256 of the archived text
31a57fa2df1c99a508bc29c502ff6e0f01123767003fb4c14ef800b50c1b7c9c- Stable URL of this page
https://metraipolis.com/documento/dc_ccc82f18d892
How to cite
Fixed-Dimensional Latent Flow for Generating Variable-Size 3D Molecules — arXiv:2609.08333 — arXiv (en), https://arxiv.org/abs/2609.08333 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ccc82f18d892 (Metraipolis archive dc_ccc82f18d892, sha256 31a57fa2df1c99a508bc29c502ff6e0f01123767003fb4c14ef800b50c1b7c9c)
BibTeX
@misc{dcccc82f18d892,
title = {Fixed-Dimensional Latent Flow for Generating Variable-Size 3D Molecules — arXiv:2609.08333},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.08333},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ccc82f18d892; sha256 31a57fa2df1c99a508bc29c502ff6e0f01123767003fb4c14ef800b50c1b7c9c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.08333. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Molecular size is coupled to composition, structure, and function, yet most 3D molecular generators require a predefined atom count. We introduce Equivariant-Free Transformer-Autoencoded Latent Flow Matching, a two-stage framework that samples a fixed-dimensional latent vector using flow matching and uses an autoregressive Transformer to determine molecular size, atom types, coordinates, and chemical attributes. Canonical atom ordering and rigid-pose alignment enable Transformers without equivariant layers, while decoded attributes guide bond reconstruction. On PCQM4Mv2, unconditional generation yields 87.9\% unique, novel molecules passing sanitization and PoseBusters checks, exceeding baselines with lower end-to-end training and sampling time and higher end-to-end throughput. Across ten target HOMO-LUMO gaps, internal ranking retains 30\% of screened candidates and increases the density functional theory-verified hit rate within 0.1 eV from 25.0\% to 52.4\%, while largely preserving novelty and diversity. These results demonstrate fixed-dimensional latent generation with autoregressive decoding as a practical approach to molecular design without prespecifying size. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.LG). Categorie arXiv: Chemical Physics (physics.chem-ph) ; Machine Learning (cs.LG); Quantitative Methods (q-bio.QM). Pagina del lavoro: https://arxiv.org/abs/2609.08333 — 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 2026-10-16T13:39:43.272Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).