{
  "abstract": "Retrosynthesis prediction is a central task in computer-aided synthesis planning, and synthesizability remains a persistent bottleneck in AI-driven molecular and materials discovery. We introduce RetroMol, a family of compact text-based models that transfers an energy-based critic’s ranking of candidate reactant sets into the generator itself. RetroBase, an 11M-parameter transformer encoderdecoder trained on two-reactant USPTO reactions, achieves Top-1 and Top-5 accuracies of 46.8% and 67.8% on a held-out test split. RetroERM, an energybased model over molecular fingerprint counts, scores RetroBase’s candidates, and its ranking preferences are distilled back into the generator. The distilled model, RetroDistil, reaches 52.6% Top-1 and 73.5% Top-5 under identical data and evaluation, improving on RetroBase by 5.7–6.8 points at Top-1/3/5 with no reranking stage at inference and no increase in model size. Distilling a critic is therefore a practical way to improve a compact retrosynthesis generator without larger or more complicated architectures.",
  "abstract_shown": true,
  "affiliations": [
    "Dept. of Chemistry & Chemical Biology",
    "McMaster University",
    "Dept. of Applied Science & Engineering",
    "University of Toronto"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 3,
  "authors": [
    {
      "affs": [
        0,
        1
      ],
      "name": "Rudra Sondhi",
      "orcid": "0009-0003-3034-7347"
    },
    {
      "affs": [
        2,
        3
      ],
      "name": "Edwin Chacko",
      "orcid": "0009-0006-0101-3091"
    },
    {
      "affs": [
        1
      ],
      "name": "Rodrigo A. Vargas-Hernández",
      "orcid": "0000-0002-5559-6521"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/53b180pzmmham.bib",
  "cited_by": 0,
  "doi": "10.26434/chemrxiv.15010571/v1",
  "fulltext_url": null,
  "id": "53b180pzmmham",
  "issns": [],
  "landing_url": "https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010571/v1",
  "language": null,
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "1703|Computational Drug Discovery Methods"
    ]
  ],
  "openalex": "W7222440616",
  "pdf_url": "https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15010571/v1",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-16",
  "publisher": "American Chemical Society (ACS)",
  "ris": "https://research.draconitic.com/paper/53b180pzmmham.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:17:24Z",
      "native_id": "10.26434/chemrxiv.15010571",
      "source": "chemrxiv",
      "url": "https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010571/v1"
    }
  ],
  "subjects": [
    {
      "field": "chemistry",
      "origin": "source:chemrxiv",
      "subfield": "general"
    }
  ],
  "title": "RetroMol: Knowledge Distillation of Energy-Based Retrosynthesis Models",
  "title_text": "RetroMol: Knowledge Distillation of Energy-Based Retrosynthesis Models",
  "type": "preprint",
  "updated_at": "2026-10-30T12:17:24Z",
  "url": "https://research.draconitic.com/paper/53b180pzmmham",
  "venue": "ChemRxiv",
  "versions": []
}