Metraipolis › archive › dc_92beec83593d

Staged Continual Adaptation of Multimodal Foundation Models for Japanese Financial Documents — arXiv:2610.27904

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.27904
License (as archived)
CC0-1.0
Archived
2026-10-21T07:06:54.574Z
SHA-256 of the archived text
e98dfb6872d6c63710ed429849af272ae74225696e334ffb9dff450b2990e234
Stable URL of this page
https://metraipolis.com/documento/dc_92beec83593d

How to cite

Staged Continual Adaptation of Multimodal Foundation Models for Japanese Financial Documents — arXiv:2610.27904 — arXiv (en), https://arxiv.org/abs/2610.27904 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_92beec83593d (Metraipolis archive dc_92beec83593d, sha256 e98dfb6872d6c63710ed429849af272ae74225696e334ffb9dff450b2990e234)
BibTeX
@misc{dc92beec83593d,
  title = {Staged Continual Adaptation of Multimodal Foundation Models for Japanese Financial Documents — arXiv:2610.27904},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.27904},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_92beec83593d; sha256 e98dfb6872d6c63710ed429849af272ae74225696e334ffb9dff450b2990e234},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.27904. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Staged post-training of multimodal foundation models chains vision-language alignment, reasoning distillation, and domain tuning, yet how each phase trades capabilities against the others remains under-characterised. Tracking an 8.4B-parameter model from an un-fine-tuned baseline through three training phases on Japanese financial documents (evaluated on five text-input benchmarks), we find the phases interact non-additively: math-only distillation overwrites the financial knowledge acquired earlier even as procedural math skill survives, and the domain phase repairs more than it installs. A skip-phase ablation is best for domain knowledge but forfeits numerical-reasoning gains, so per-phase contributions are not decomposable and the best checkpoint is task-dependent, with implications for checkpoint retention and scheduling rather than added stages.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.27904 — 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 2026-10-21T07:05:43.452Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).