Metraipolis › archive › dc_4ec49ea661af

Dynamic Alignment and Calibration for Multimodal Learning — arXiv:2610.07928

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.07928
License (as archived)
CC0-1.0
Archived
2026-10-07T11:21:21.925Z
SHA-256 of the archived text
e521760d724e54f24bc778548650afe224ee1b8177313b6ee49143c03ec98f8e
Stable URL of this page
https://metraipolis.com/documento/dc_4ec49ea661af

How to cite

Dynamic Alignment and Calibration for Multimodal Learning — arXiv:2610.07928 — arXiv (en), https://arxiv.org/abs/2610.07928 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4ec49ea661af (Metraipolis archive dc_4ec49ea661af, sha256 e521760d724e54f24bc778548650afe224ee1b8177313b6ee49143c03ec98f8e)
BibTeX
@misc{dc4ec49ea661af,
  title = {Dynamic Alignment and Calibration for Multimodal Learning — arXiv:2610.07928},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.07928},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4ec49ea661af; sha256 e521760d724e54f24bc778548650afe224ee1b8177313b6ee49143c03ec98f8e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.07928. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Dynamic multimodal learning aims to learn robust representations by adaptively modeling information discrepancies across modalities. However, existing methods still suffer from two limitations: (i) static cross-modal alignment strategies usually impose uniform constraints on all samples while overlooking sample-wise variations, potentially leading to unreasonable over-alignment; and (ii) confidence- or uncertainty-aware fusion methods often fail to adequately account for feature magnitude and confidence differences across modalities. For modality pairs with significant feature magnitude differences or small confidence gaps, it might be unreliable to strictly align fusion weights according to confidence. To address these issues, we propose an Alignment- and Calibration-driven Multimodal Learning framework (ACML). Specifically, ACML incorporates a dynamic cross-modal triplet alignment module, which enforces strong semantic consistency for high-confidence positive pairs while encouraging diverse representation learning between high- and low-confidence positive pairs according to their confidence gaps. Additionally, ACML introduces a difference-aware attention calibration strategy that adaptively adjusts attention regularization based on feature magnitude and confidence differences across modalities, thereby mitigating biases caused by unreasonable fusion constraints. Extensive experiments on multiple multimodal benchmark datasets demonstrate that ACML consistently achieves superior performance and robustness over recent state-of-the-art methods.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.07928 — 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 2026-10-07T11:15:24.784Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).