{
  "abstract": "Self-distillation methods for large language models with an exponential moving average (EMA) teacher already retain a history of student updates. Muon nevertheless stores a separate momentum buffer, increasing persistent training memory. We propose Muon with Reused Teacher (Muon-RT), which reuses the teacher-student displacement to eliminate momentum buffers for hidden weight matrices. We show that this displacement encodes an exponentially weighted history of applied updates and smooths temporally uncorrelated noise under a local model. Muon-RT removes its radial component, normalizes it spectrally, and mixes it with the current gradient after Newton-Schulz (NS) orthogonalization. Under explicit regularity and alignment assumptions, we establish linear contraction of an expected objective-gap bound toward a residual error level. With Self-Distillation Fine-Tuning (SDFT) on Qwen2.5-7B-Instruct, Muon-RT improves average accuracy over AdamW by 2.6 points with 42% less persistent memory, and over momentum Muon by 5.4 points with 27% less persistent memory. With Self-Distillation Policy Optimization (SDPO) on Olmo-3-7B-Instruct, it improves average accuracy over AdamW by 2.0 points.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.41104",
  "author_total": 3,
  "authors": [
    {
      "name": "Aryan Das"
    },
    {
      "name": "Vinay Kumar Verma"
    },
    {
      "name": "Aakansha Mishra"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/e7c0vr0qymhrj.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.41104",
  "fulltext_url": null,
  "id": "e7c0vr0qymhrj",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.41104",
  "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": null,
  "pdf_url": "https://arxiv.org/pdf/2610.41104",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-30",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/e7c0vr0qymhrj.ris",
  "sources": [
    {
      "first_seen": "2026-11-02T12:21:25Z",
      "native_id": "2610.41104",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.41104"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "MUON-RT Reusing the Teacher's Trajectory for Momentum-Free Muon for LLM Self-Distillation",
  "title_text": "MUON-RT Reusing the Teacher's Trajectory for Momentum-Free Muon for LLM Self-Distillation",
  "type": "preprint",
  "updated_at": "2026-11-04T15:20:53Z",
  "url": "https://research.draconitic.com/paper/e7c0vr0qymhrj",
  "venue": "arXiv",
  "versions": []
}