Metraipolis › archive › dc_b3135ca23746

Multimodal Information Interactions Bottleneck for Multimodal Self-Supervised Learning — arXiv:2611.07718

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.07718
License (as archived)
CC0-1.0
Archived
2026-11-06T14:51:14.137Z
SHA-256 of the archived text
d7ced6e8f93243aa42b793a1983fbecabbf6992ea4c06ea184ced7602efa1390
Stable URL of this page
https://metraipolis.com/documento/dc_b3135ca23746

How to cite

Multimodal Information Interactions Bottleneck for Multimodal Self-Supervised Learning — arXiv:2611.07718 — arXiv (en), https://arxiv.org/abs/2611.07718 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b3135ca23746 (Metraipolis archive dc_b3135ca23746, sha256 d7ced6e8f93243aa42b793a1983fbecabbf6992ea4c06ea184ced7602efa1390)
BibTeX
@misc{dcb3135ca23746,
  title = {Multimodal Information Interactions Bottleneck for Multimodal Self-Supervised Learning — arXiv:2611.07718},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.07718},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b3135ca23746; sha256 d7ced6e8f93243aa42b793a1983fbecabbf6992ea4c06ea184ced7602efa1390},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.07718. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multimodal self-supervised learning is challenged by heterogeneity, unequal informativeness, and complex information interactions across multimodal data sources such as images, text, audio, and tabular data. While contrastive methods effectively align shared semantics, they typically rely on strong multi-view redundancy assumptions, overlooking modality-specific and synergistic information, thus leading to suboptimal representations. We propose the Multimodal Information Interactions Bottleneck (MIIB), a principled framework grounded in Partial Information Decomposition that explicitly encodes redundant, unique, and synergistic information within multimodal interactions. MIIB employs a dual-path model architecture for modality-specific encoding and multimodal fusion, optimized using surrogate tractable entropy-based objectives that enforce minimal sufficient representations without requiring explicit mutual information estimation. By leveraging variational bounds and score-based entropy estimation, MIIB efficiently captures complex cross-modal interactions while suppressing superfluous information. Experiments across diverse benchmarks demonstrate consistent improvements over state-of-the-art multimodal self-supervised methods. Code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.07718 — 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 2026-11-06T14:21:42.663Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).