{
  "abstract": "Mixture-of-Experts (MoE) layers are central to frontier language models, and recent architectures push toward more and smaller experts. In this regime, the standard linear router becomes a bottleneck: with $M$ experts and hidden dimension $h$, its per-token cost $Θ(Mh)$ dominates the MoE layer once $M$ is large. We introduce MoRE (Mixture of Rank-reduced-routed Experts), which factorizes the router weight matrix at rank $r$ and reduces the routing cost to $O((h + M)r)$. We prove that rank logarithmic in $M$ suffices for routing expressivity when the number of active experts is fixed, and is necessary up to precision factors. We also prove that logarithmic rank preserves load balance in a Gaussian memorization model, and training on a synthetic phonebook task shows that low rank does not hurt memorization. At matched active FLOPs, the factorization allows a factor of $Θ(h/r)$ more experts. To realize this gain in wall-clock time, we design a fused Triton kernel at inference that avoids expensive memory operations on HBM. Empirically, MoRE improves memorization on the phonebook task and performance on knowledge-intensive Q\\&A benchmarks after pretraining, while matching reasoning ability. Code available at https://github.com/Matheart/MoRE_code.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.36301",
  "author_total": 3,
  "authors": [
    {
      "name": "Honam Wong"
    },
    {
      "name": "Surbhi Goel"
    },
    {
      "name": "Enric Boix-Adserà"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/dn6xeavewmz8w.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.36301",
  "fulltext_url": null,
  "id": "dn6xeavewmz8w",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.36301",
  "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.AI"
    ],
    [
      "arxiv",
      "cs.CL"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1707|Advanced Neural Network Applications"
    ]
  ],
  "openalex": "W7214947677",
  "pdf_url": "https://arxiv.org/pdf/2609.36301",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-28",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/dn6xeavewmz8w.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:25:41Z",
      "native_id": "2609.36301",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.36301"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "MoRE: Scaling mixture of experts with hardware-aware low-rank routing",
  "title_text": "MoRE: Scaling mixture of experts with hardware-aware low-rank routing",
  "type": "preprint",
  "updated_at": "2026-10-30T12:25:41Z",
  "url": "https://research.draconitic.com/paper/dn6xeavewmz8w",
  "venue": "arXiv",
  "versions": []
}