mlsanityData to 14 Oct 2026

Robust Multimodal Sentiment Analysis via Double Information Bottleneck

arXiv 2511.01444 ↗First posted 3 Nov 2025cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Multimodal sentiment analysis has received significant attention across diverse research domains. Despite advancements in algorithm design, existing approaches suffer from two critical limitations: insufficient learning of noise-contaminated unimodal data, leading to corrupted cross-modal interactions, and inadequate fusion of multimodal representations, resulting in discarding discriminative unimodal information while retaining multimodal redundant information. To address these challenges, this paper proposes a Double Information Bottleneck (DIB) strategy to obtain a powerful, unified compact multimodal representation. Implemented within the framework of low-rank Renyi's entropy functional, DIB offers enhanced robustness against diverse noise sources and computational tractability for high-dimensional data, as compared to the conventional Shannon entropy-based methods. The DIB comprises two key modules: 1) learning a sufficient and compressed representation of individual unimodal data by maximizing the task-relevant information and discarding the superfluous information, and 2) ensuring the discriminative ability of multimodal representation through a novel attention bottleneck fusion mechanism. Consequently, DIB yields a multimodal representation that effectively filters out noisy information from unimodal data while capturing inter-modal complementarity. Extensive experiments on CMU-MOSI, CMU-MOSEI, CH-SIMS, and MVSA-Single validate the effectiveness of our method. The model achieves 47.4% accuracy under the Acc-7 metric on CMU-MOSI and 81.63% F1-score on CH-SIMS, outperforming the second-best baseline by 1.19%. Under noise, it shows only 0.36% and 0.29% performance degradation on CMU-MOSI and CMU-MOSEI respectively.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 37 of its 94 references that are in the map. First posted on arXiv 3 Nov 2025.

Builds on

37 of 94 references.

References · 37

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2016Variational Inference: A Review for Statisticianscited by 1,472
  3. 2016Deep Variational Information Bottleneckcited by 930
  4. 2019EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Taskscited by 807
  5. 2015Deep Learning and the Information Bottleneck Principlecited by 802
  6. 2017AttnGAN: Fine-Grained Text to Image Generation with Attentional Generative Adversarial Networkscited by 664
  7. 2019Multimodal Transformer for Unaligned Multimodal Language Sequencescited by 645
  8. 2017Opening the Black Box of Deep Neural Networks via Informationcited by 503
  9. 2017Tensor Fusion Network for Multimodal Sentiment Analysiscited by 433
  10. 2021Attention Bottlenecks for Multimodal Fusioncited by 361
  11. 2018Efficient Low-rank Multimodal Fusion with Modality-Specific Factorscited by 270
  12. 2020MISA: Modality-Invariant and -Specific Representations for Multimodal Sentiment Analysiscited by 265
25 more
  1. 2016MOSI: Multimodal Corpus of Sentiment Intensity and Subjectivity Analysis in Online Opinion Videoscited by 194
  2. 2021Learning Modality-Specific Representations with Self-Supervised Multi-Task Learning for Multimodal Sentiment Analysiscited by 180
  3. 2019Towards Multimodal Sarcasm Detection (An _Obviously_ Perfect Paper)cited by 174
  4. 2018Memory Fusion Network for Multi-view Sequential Learningcited by 168
  5. 2023A Survey of Large Language Models for Healthcare: from Data, Technology, and Applications to Accountability and Ethicscited by 163
  6. 2016Information Dropout: Learning Optimal Representations Through Noisy Computationcited by 132
  7. 2021Improving Multimodal Fusion with Hierarchical Mutual Information Maximization for Multimodal Sentiment Analysiscited by 131
  8. 2018Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flowcited by 105
  9. 2019InfoBot: Transfer and Exploration via the Information Bottleneckcited by 78
  10. 2018Learning Representations for Neural Network-Based Classification Using the Information Bottleneck Principlecited by 70
  11. 2022Multimodal Information Bottleneck: Learning Minimal Sufficient Unimodal and Multimodal Representationscited by 48
  12. 2019Modality to Modality Translation: An Adversarial Representation Learning and Graph Fusion Network for Multimodal Fusioncited by 45
  13. 2021Hybrid Contrastive Learning of Tri-Modal Representation for Multimodal Sentiment Analysiscited by 41
  14. 2023Learning Language-guided Adaptive Hyper-modality Representation for Multimodal Sentiment Analysiscited by 29
  15. 2018Multi-task Learning of Hierarchical Vision-Language Representationcited by 26
  16. 2020MSAF: Multimodal Split Attention Fusioncited by 23
  17. 2022CLMLF:A Contrastive Learning and Multi-Layer Fusion Method for Multimodal Sentiment Detectioncited by 21
  18. 2022Efficient Multimodal Transformer with Dual-Level Feature Restoration for Robust Multimodal Sentiment Analysiscited by 20
  19. 2024Has Multimodal Learning Delivered Universal Intelligence in Healthcare? A Comprehensive Surveycited by 20
  20. 2017Benchmarking Multimodal Sentiment Analysiscited by 11
  21. 2021Graph Capsule Aggregation for Unaligned Multimodal Sequencescited by 6
  22. 2021Multi-Modal Mutual Information Maximization: A Novel Approach for Unsupervised Deep Cross-Modal Hashingcited by 4
  23. 2022Holistic Visual-Textual Sentiment Analysis with Prior Modelscited by 3
  24. 2024Meta-Learn Unimodal Signals with Weak Supervision for Multimodal Sentiment Analysiscited by 2
  25. 2022Robust and Fast Measure of Information via Low-rank Representationcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Multi-Granularity Sentiment Integration for LLM-Based Multimodal Sentiment Analysiscited by 0