Metraipolis › archive › dc_2b0d520dff34

Breaking Information Islands in Sparse Tuning via Small-World Connectivity — arXiv:2610.41665

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41665
License (as archived)
CC0-1.0
Archived
2026-11-03T07:21:56.966Z
SHA-256 of the archived text
044e25e68bf05a4061fb8c4a3647086d7456cf9263b7dba0245b4b40f7bf1528
Stable URL of this page
https://metraipolis.com/documento/dc_2b0d520dff34

How to cite

Breaking Information Islands in Sparse Tuning via Small-World Connectivity — arXiv:2610.41665 — arXiv (en), https://arxiv.org/abs/2610.41665 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2b0d520dff34 (Metraipolis archive dc_2b0d520dff34, sha256 044e25e68bf05a4061fb8c4a3647086d7456cf9263b7dba0245b4b40f7bf1528)
BibTeX
@misc{dc2b0d520dff34,
  title = {Breaking Information Islands in Sparse Tuning via Small-World Connectivity — arXiv:2610.41665},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41665},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2b0d520dff34; sha256 044e25e68bf05a4061fb8c4a3647086d7456cf9263b7dba0245b4b40f7bf1528},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41665. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Sparse tuning is widely used to adapt large language models due to its parameter efficiency. However, its effectiveness depends not only on the number of trainable parameters, but also on the feature-interaction topology of the sparse update's structural support. Limited or uneven connectivity in this support may isolate subsets of feature dimensions, forming information islands that restrict direct task-specific interactions through the trainable update. To address this issue, we propose Small-World Induced Fine-Tuning (SWIFT), a sparse tuning framework inspired by small-world connectivity. SWIFT combines block-diagonal sparse updates for local feature mixing with a fixed deterministic scattering permutation that routes the mixed outputs across feature groups without additional trainable parameters. We further show theoretically that certain random sparse-update models and block-diagonal updates can exhibit information islands, whereas a sufficiently scattering permutation guarantees positive block-level expansion when the trainable block-diagonal factor consists of equal-sized, structurally dense square blocks. Extensive experiments on commonsense reasoning, natural language generation, and image classification demonstrate that SWIFT outperforms the evaluated parameter-efficient fine-tuning (PEFT) baselines on average and can match or surpass full fine-tuning. Implementation is available at this https URL

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.41665 — 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 2026-11-03T07:19:48.835Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).