EUVIMEDEUVIMED entdecken
Crossref

Aligning the True Semantics: Constrained Decoupling and Distribution Sampling for Cross-Modal Alignment

Xiang Ma, Lexin Fang, Litian Xu, Caiming Zhang

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (29) · 24290-24298

DOI: 10.1609/aaai.v40i29.39609 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Xiang Ma und 3 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cross-modal alignment is a crucial task in multimodal learning aimed at achieving semantic consistency between vision and language. This requires that image-text pairs exhibit similar semantics. Traditional algorithms pursue embedding consistency to achieve semantic consistency, ignoring the non-semantic information present in the embedding. An intuitive approach is to decouple the embeddings into semantic and modality components, aligning only the semantic component. However, this introduces two main challenges: (1) There is no established standard for distinguishing semantic and modal information. (2) The modality gap can cause semantic alignment deviation or information loss. To align the true semantics, we propose a novel cross-modal alignment algorithm via Constrained Decoupling and Distribution Sampling (CDDS). Specifically, (1) A dual-path UNet is introduced to adaptively decouple the embeddings, applying multiple constraints to ensure effective separation. (2) A distribution sampling method is proposed to bridge the modality gap, ensuring the rationality of the alignment process. Extensive experiments on various benchmarks and model backbones demonstrate the superiority of CDDS, outperforming state-of-the-art methods by 6.6% to 14.2%.

Zitieren

APA

Ma, X., Fang, L., Xu, L., & Zhang, C. (2026). Aligning the True Semantics: Constrained Decoupling and Distribution Sampling for Cross-Modal Alignment. Proceedings of the AAAI Conference on Artificial Intelligence, 40(29), 24290-24298. https://doi.org/10.1609/aaai.v40i29.39609

Vancouver

Ma X, Fang L, Xu L, Zhang C. Aligning the True Semantics: Constrained Decoupling and Distribution Sampling for Cross-Modal Alignment. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(29):24290-24298. doi:10.1609/aaai.v40i29.39609

BibTeX
@article{Ma2026,
  author = {Ma, Xiang and Fang, Lexin and Xu, Litian and Zhang, Caiming},
  title = {Aligning the True Semantics: Constrained Decoupling and Distribution Sampling for Cross-Modal Alignment},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {29},
  pages = {24290--24298},
  doi = {10.1609/aaai.v40i29.39609}
}
Kostenlos starten →