Metraipolis › archive › dc_7062a82dbdb3
Continual Adaptation Across Modalities: Learning and Forgetting in Unified Multimodal Models — arXiv:2610.38436
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.38436
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-01T02:24:48.188Z
- SHA-256 of the archived text
5beeeb57f526cd12b3f1d74ce47aba99964f78897f5f1aea455b60811e642bed- Stable URL of this page
https://metraipolis.com/documento/dc_7062a82dbdb3
How to cite
Continual Adaptation Across Modalities: Learning and Forgetting in Unified Multimodal Models — arXiv:2610.38436 — arXiv (en), https://arxiv.org/abs/2610.38436 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7062a82dbdb3 (Metraipolis archive dc_7062a82dbdb3, sha256 5beeeb57f526cd12b3f1d74ce47aba99964f78897f5f1aea455b60811e642bed)
BibTeX
@misc{dc7062a82dbdb3,
title = {Continual Adaptation Across Modalities: Learning and Forgetting in Unified Multimodal Models — arXiv:2610.38436},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.38436},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7062a82dbdb3; sha256 5beeeb57f526cd12b3f1d74ce47aba99964f78897f5f1aea455b60811e642bed},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.38436. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Unified multimodal models (UMMs), which can process and generate diverse modalities within a single framework, have recently emerged as a powerful paradigm. However, when these models undergo continual learning on multiple tasks across different modalities, their learning and forgetting dynamics remain poorly understood. In this paper, we systematically investigate how architectural design and adaptation method shape intra-modal learning and cross-modal interference between the generation and understanding sides. Specifically, we report three findings. First, the severity of cross-modal interference decreases with the degree of modality-specific parameter isolation. Second, a within-model asymmetry: the generation side of a UMM is markedly more fragile than its understanding side. Third, what post-training adds is the first to be forgotten: purely post-training capabilities such as personalized generation collapse by 84\% to 90\%, while pretraining capabilities remain stable. Together, these results clarify how parameter isolation and post-training fragility shape UMMs' learning and forgetting dynamics, offering a starting point for developing continually evolving UMMs. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.38436 — 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 2026-11-01T02:12:44.122Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).