{
  "abstract": "Genome foundation models are growing into sparse mixture-of-experts (MoE) networks whose expert weights no longer fit on the machines that hold the sequences, yet sending a private genome to rented accelerators exposes it: we show that a single server hosting one expert recovers the input nucleotides with 99.8% top-1 accuracy. We present CipherGenome, a protocol that keeps the embedding, attention and router of a 15.1B-parameter MoE genome model on a trusted thin client and outsources every expert projection, 95.8% of the parameters, to untrusted and possibly colluding GPU servers under module-LWE encryption. The design exploits three structural facts: expert layers are linear between two SwiGLU gates, expert weights are public, and GPU integer tensor cores can evaluate a ciphertext-weight product exactly modulo $2^{48}$ in a single GEMM. The client evaluates the nonlinearity exactly and re-encrypts with fresh secrets, so no polynomial approximation or bootstrapping is ever needed. On 72 windows from 12 bacterial genomes, encryption adds $2.54 \\times 10^{-4}$ nats per token of KL divergence (95% CI upper bound $3.95 \\times 10^{-4}$), below a pre-registered non-inferiority margin and indistinguishable from bf16 inference, while the same inversion attack falls to chance level. A reusable public hint cuts end-to-end latency by 3.54 times, wire compression reduces traffic 6.8 times, per-layer padding reduces routing leakage from 54.9% to 8.9% accuracy, and HE-compatible int4 experts remain non-inferior to their plaintext counterparts. Per expert and token, the server-side cost is more than six orders of magnitude below a CKKS baseline.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.35883",
  "author_total": 2,
  "authors": [
    {
      "name": "Guang Yang"
    },
    {
      "name": "Fengchen Liu"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/7p2dz69veypg3.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.35883",
  "fulltext_url": null,
  "id": "7p2dz69veypg3",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.35883",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.CR"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "q-bio.GN"
    ],
    [
      "openalex",
      "1702|Cryptography and Data Security"
    ]
  ],
  "openalex": "W7214920434",
  "pdf_url": "https://arxiv.org/pdf/2609.35883",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-01",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/7p2dz69veypg3.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:26:50Z",
      "native_id": "2609.35883",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.35883"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "arxiv:q-bio.GN",
      "subfield": "genetics-genomics"
    }
  ],
  "title": "CipherGenome: Homomorphic Inference for Genomic Mixture-of-Experts",
  "title_text": "CipherGenome: Homomorphic Inference for Genomic Mixture-of-Experts",
  "type": "preprint",
  "updated_at": "2026-10-30T12:26:50Z",
  "url": "https://research.draconitic.com/paper/7p2dz69veypg3",
  "venue": "arXiv",
  "versions": []
}