Metraipolis › archive › dc_d538fe7892e4

SaMA: Morpho Adaptation via Asymmetric Expansion of Kronecker Product — arXiv:2611.04462

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04462
License (as archived)
CC0-1.0
Archived
2026-11-04T12:29:55.755Z
SHA-256 of the archived text
eb5a71b60e217fda4eefc1b1387afd8a690167b139a06265445e5d7ac8f6581f
Stable URL of this page
https://metraipolis.com/documento/dc_d538fe7892e4

How to cite

SaMA: Morpho Adaptation via Asymmetric Expansion of Kronecker Product — arXiv:2611.04462 — arXiv (en), https://arxiv.org/abs/2611.04462 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d538fe7892e4 (Metraipolis archive dc_d538fe7892e4, sha256 eb5a71b60e217fda4eefc1b1387afd8a690167b139a06265445e5d7ac8f6581f)
BibTeX
@misc{dcd538fe7892e4,
  title = {SaMA: Morpho Adaptation via Asymmetric Expansion of Kronecker Product — arXiv:2611.04462},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04462},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d538fe7892e4; sha256 eb5a71b60e217fda4eefc1b1387afd8a690167b139a06265445e5d7ac8f6581f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04462. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Low-rank adaptation (LoRA) has attracted significant attention in the parameter-efficient fine-tuning (PEFT) landscape. However, its design suffers from two fundamental limitations. First, its invariance manifold induces flat directions in the loss landscape, creating an optimization bottleneck for adaptive optimizers. Second, its effective rank scales only linearly with the parameter budget. In this paper, we propose Scalable Morpho Adaptation (SaMA), a PEFT method grounded in a principled generalization of the Kronecker product. By relaxing the shared-block constraint in the perfect-shuffle factorization, SaMA spans from low to full rank and achieves a practical rank that scales quadratically with the parameter budget, up to full-rank. Furthermore, we show that its invariance manifold collapses to a diagonal group, making it structurally more optimizer-friendly. Empirically, SaMA consistently outperforms strong PEFT baselines on commonsense and arithmetic reasoning benchmarks while being more parameter-efficient.

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).
Pagina del lavoro: https://arxiv.org/abs/2611.04462 — 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 2026-11-04T12:26:47.277Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).