Metraipolis › archive › dc_fdf71b2aecae

MUON-RT Reusing the Teacher's Trajectory for Momentum-Free Muon for LLM Self-Distillation — arXiv:2610.41104

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.41104
License (as archived): CC0-1.0
Archived: 2026-11-02T04:44:35.966Z
SHA-256 of the archived text: 68e589fd764d9770d996e9dd211b2abf1b99b4932ba59025cbb5c954696485d4
Stable URL of this page: https://metraipolis.com/documento/dc_fdf71b2aecae

How to cite

MUON-RT Reusing the Teacher's Trajectory for Momentum-Free Muon for LLM Self-Distillation — arXiv:2610.41104 — arXiv (en), https://arxiv.org/abs/2610.41104 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fdf71b2aecae (Metraipolis archive dc_fdf71b2aecae, sha256 68e589fd764d9770d996e9dd211b2abf1b99b4932ba59025cbb5c954696485d4)

BibTeX

@misc{dcfdf71b2aecae,
  title = {MUON-RT Reusing the Teacher's Trajectory for Momentum-Free Muon for LLM Self-Distillation — arXiv:2610.41104},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41104},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fdf71b2aecae; sha256 68e589fd764d9770d996e9dd211b2abf1b99b4932ba59025cbb5c954696485d4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41104. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.41104 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-02T04:42:12.447Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).