Metraipolis › archive › dc_dc734c280d0d

CAMFT: Conflict-Aware Mergeable Fine-Tuning for Large Language Models — arXiv:2609.22253

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.22253
License (as archived)
CC0-1.0
Archived
2026-10-09T06:50:25.761Z
SHA-256 of the archived text
2c71b61394028d1a1aa070b139304fd6fcf0ce3add52d0da727d77f1f4f511f5
Stable URL of this page
https://metraipolis.com/documento/dc_dc734c280d0d

How to cite

CAMFT: Conflict-Aware Mergeable Fine-Tuning for Large Language Models — arXiv:2609.22253 — arXiv (en), https://arxiv.org/abs/2609.22253 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dc734c280d0d (Metraipolis archive dc_dc734c280d0d, sha256 2c71b61394028d1a1aa070b139304fd6fcf0ce3add52d0da727d77f1f4f511f5)
BibTeX
@misc{dcdc734c280d0d,
  title = {CAMFT: Conflict-Aware Mergeable Fine-Tuning for Large Language Models — arXiv:2609.22253},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.22253},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dc734c280d0d; sha256 2c71b61394028d1a1aa070b139304fd6fcf0ce3add52d0da727d77f1f4f511f5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.22253. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Model merging has emerged as a promising paradigm for integrating multiple task-specific capabilities into a single large language model. However, existing methods predominantly focus on post-hoc processing of independently fine-tuned models, overlooking how the training phase itself impacts cross-task compatibility. Resolving parameter conflicts after fine-tuning is inherently sub-optimal. To address this, we propose CAMFT, a Conflict-Aware Mergeable Fine-Tuning method that makes task adaptation both efficient and mergeaware. CAMFT treats mergeability as a property shaped during fine-tuning, rather than only a problem to be solved after fine-tuning. By guiding each task to update sparse coordinates with lower cross-task conflict, CAMFT produces task updates that are efficient to train and more compatible for downstream model merging. Extensive experiments demonstrate that CAMFT outperforms standard finetuning baselines in multi-task merging scenarios. Codes are available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CL).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.22253 — 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 2026-10-09T06:48:45.099Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).