{
  "abstract": "Molecular size is intrinsically linked to composition, structure, and properties, but most 3D molecular generators treat size as a predefined input rather than a quantity to be generated. We introduce Equivariant-Free Transformer-Autoencoded Latent Flow Matching, a two-stage framework that samples a fixed-dimensional latent vector using flow matching and employs 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 and 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 the screened candidates and increases the density functional theory-verified hit rate within 0.1 eV of the target 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.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.08333",
  "author_total": 4,
  "authors": [
    {
      "name": "Weichi Yao"
    },
    {
      "name": "Cameron Gruich"
    },
    {
      "name": "Bryan R. Goldsmith"
    },
    {
      "name": "Yixin Wang"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/9xqy23ydg3c2e.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.08333",
  "fulltext_url": null,
  "id": "9xqy23ydg3c2e",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.08333",
  "language": null,
  "licence": "LicenseRef-arXiv-nonexclusive-1.0",
  "licence_label": "arXiv licence",
  "licence_source": "article",
  "licence_url": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "physics.chem-ph"
    ],
    [
      "arxiv",
      "q-bio.QM"
    ],
    [
      "openalex",
      "1703|Computational Drug Discovery Methods"
    ]
  ],
  "openalex": "W7212001875",
  "pdf_url": "https://arxiv.org/pdf/2609.08333",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-21",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/9xqy23ydg3c2e.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:34:34Z",
      "native_id": "2609.08333",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.08333"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "arxiv:q-bio.QM",
      "subfield": "bioinformatics"
    },
    {
      "field": "chemistry",
      "origin": "arxiv:physics.chem-ph",
      "subfield": "physical"
    }
  ],
  "title": "Fixed-Dimensional Latent Flow for Generating Variable-Size 3D Molecules",
  "title_text": "Fixed-Dimensional Latent Flow for Generating Variable-Size 3D Molecules",
  "type": "preprint",
  "updated_at": "2026-10-30T12:34:38Z",
  "url": "https://research.draconitic.com/paper/9xqy23ydg3c2e",
  "venue": "arXiv",
  "versions": []
}