{
  "abstract": "Mathematical invariants provide complementary representations of molecular structure, but their predictive utility can vary across datasets and prediction tasks. For a new dataset, appropriate construction of invariant representations from algebraic topology, differential geometry, topological spectral theory, and commutative algebra is a challenging task, particularly when evaluating multiple representations is computationally expensive. We develop a Mathematical Agentic Model (MathAgent) to automatically optimize invariant representations in molecular property prediction. The framework combines dataset-aware representation design, low-cost candidate screening, and progressive evaluation. Language-model reasoning supports task interpretation, strategy proposal, and workflow coordination, while scientific computations and numerical decisions are carried out by deterministic tools. We evaluate MathAgent on protein--ligand binding-affinity and quantitative toxicity prediction tasks. Different datasets favor different combinations of mathematical invariants, demonstrating the need for dataset-adaptive representation selection rather than a fixed universal representation. Objective-conditioned executions further show that the framework can adjust its evaluation strategy according to user-specified goals. MathAgent therefore provides a systematic, evidence-driven, and traceable approach for optimizing mathematical invariant representations and reducing reliance on manual configuration and/or experimentation in molecular property prediction.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.03992",
  "author_total": 3,
  "authors": [
    {
      "name": "Yiming Ren"
    },
    {
      "name": "Josh V. Vermaas"
    },
    {
      "name": "Guo-Wei Wei"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/1bwych66z60wb.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.03992",
  "fulltext_url": null,
  "id": "1bwych66z60wb",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.03992",
  "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",
      "q-bio.BM"
    ],
    [
      "openalex",
      "1703|Computational Drug Discovery Methods"
    ]
  ],
  "openalex": "W7220695063",
  "pdf_url": "https://arxiv.org/pdf/2610.03992",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-02",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/1bwych66z60wb.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:11Z",
      "native_id": "2610.03992",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.03992"
    }
  ],
  "subjects": [
    {
      "field": "biochemistry",
      "origin": "arxiv:q-bio.BM",
      "subfield": "biophysics"
    }
  ],
  "title": "MathAgent: Multi-Agent Optimization of Mathematical Invariants for Molecular Property Prediction",
  "title_text": "MathAgent: Multi-Agent Optimization of Mathematical Invariants for Molecular Property Prediction",
  "type": "preprint",
  "updated_at": "2026-10-30T12:27:11Z",
  "url": "https://research.draconitic.com/paper/1bwych66z60wb",
  "venue": "arXiv",
  "versions": []
}