Bridge: A Cross-Modal Learning Framework for Unified Semantic Representation in Noisy Communication

Chen, Liang; Huang, Yanze; Lin, Limei; Wang, Xiaoding; Lou, Wei; Wu, Jie; Hsieh, Sun-Yuan

doi:10.24963/ijcai.2026/465

Bridge: A Cross-Modal Learning Framework for Unified Semantic Representation in Noisy Communication

Liang Chen, Yanze Huang, Limei Lin, Xiaoding Wang, Wei Lou, Jie Wu, Sun-Yuan Hsieh

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 4179-4187.

https://doi.org/10.24963/ijcai.2026/465

Keywords

Machine Learning: Representation learning

Abstract

Multimodal semantic communication systems face a critical challenge in extracting and aligning semantic features across heterogeneous modalities within a unified representation space, particularly under noisy transmission conditions. To address this, we propose Bridge, a cross-modal learning framework that integrates video, audio, and text into a unified semantic space through feature disentanglement and contrastive alignment. Bridge separates modality-specific and modality-invariant representations, enhancing both intra-modal precision and inter-modal generalization. During decoding, it leverages heterogeneous foundation models to preserve semantic fidelity under channel noise. Extensive experiments on multiple multimodal benchmarks under varying SNR levels demonstrate that Bridge maintains high semantic consistency and reconstruction quality, improving image reconstruction by approximately 16.3% in low-SNR regimes. Our work provides a practical and theoretically grounded framework for next-generation multimodal semantic communication systems.

Liang Chen, Yanze Huang, Limei Lin, Xiaoding Wang, Wei Lou, Jie Wu, and Sun-Yuan Hsieh. 2026. Bridge: A Cross-Modal Learning Framework for Unified Semantic Representation in Noisy Communication. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 4179–4187. https://doi.org/10.24963/ijcai.2026/465

@inproceedings{ijcai2026p465,
  title     = {Bridge: A Cross-Modal Learning Framework for Unified Semantic Representation in Noisy Communication},
  author    = {Chen, Liang and Huang, Yanze and Lin, Limei and Wang, Xiaoding and Lou, Wei and Wu, Jie and Hsieh, Sun-Yuan},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {4179--4187},
  year      = {2026},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2026/465},
  url       = {https://doi.org/10.24963/ijcai.2026/465},
}

TY  - CPAPER
TI  - Bridge: A Cross-Modal Learning Framework for Unified Semantic Representation in Noisy Communication
AU  - Chen, Liang
AU  - Huang, Yanze
AU  - Lin, Limei
AU  - Wang, Xiaoding
AU  - Lou, Wei
AU  - Wu, Jie
AU  - Hsieh, Sun-Yuan
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 4179
EP  - 4187
DO  - 10.24963/ijcai.2026/465
UR  - https://doi.org/10.24963/ijcai.2026/465
ER  -