# Multimodal Information Interactions Bottleneck for Multimodal Self-Supervised Learning

**Authors:** Liangjian Wen, Jiangtao Zheng, Jiang Duan, Qun Dai, Yong Dai, Jianzhuang Liu, Jun Wang, Shu-Dong Huang, Xinwang Liu, Zenglin Xu, Zhao Kang  
**Lab:** Southwestern University of Finance and Economics · **Published:** 5 Nov 2026  
**Links:** [arXiv](https://arxiv.org/abs/2611.07718) · [PDF](https://arxiv.org/pdf/2611.07718) · [Code](https://github.com/wanglingzaishi/MIIB) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2611.07718.html)

## Abstract

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 https://github.com/wanglingzaishi/MIIB.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.