# Efficient Multimodal Transformer with Dual-Level Feature Restoration for Robust Multimodal Sentiment Analysis

arXiv: https://arxiv.org/abs/2208.07589 · HTML: https://mlsanity.com/paper/2208.07589/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- Builds on 40 of its 76 references that are in the map.
- First posted on arXiv 16 Aug 2022.

## Builds on (most cited first, 30 of 40)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Exploring Simple Siamese Representation Learning](https://mlsanity.com/paper/2011.10566/) (2020)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [Context Encoders: Feature Learning by Inpainting](https://mlsanity.com/paper/1604.07379/) (2016)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey](https://mlsanity.com/paper/1902.06162/) (2019)
- [Multimodal Transformer for Unaligned Multimodal Language Sequences](https://mlsanity.com/paper/1906.00295/) (2019)
- [Tensor Fusion Network for Multimodal Sentiment Analysis](https://mlsanity.com/paper/1707.07250/) (2017)
- [Attention Bottlenecks for Multimodal Fusion](https://mlsanity.com/paper/2107.00135/) (2021)
- [Efficient Low-rank Multimodal Fusion with Modality-Specific Factors](https://mlsanity.com/paper/1806.00064/) (2018)
- [MISA: Modality-Invariant and -Specific Representations for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2005.03545/) (2020)
- [Learning Modality-Specific Representations with Self-Supervised Multi-Task Learning for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2102.04830/) (2021)
- [Memory Fusion Network for Multi-view Sequential Learning](https://mlsanity.com/paper/1802.00927/) (2018)
- [Found in Translation: Learning Robust Joint Representations by Cyclic Translations Between Modalities](https://mlsanity.com/paper/1812.07809/) (2018)
- [Improving Multimodal Fusion with Hierarchical Mutual Information Maximization for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2109.00412/) (2021)
- [Multi-attention Recurrent Network for Human Communication Comprehension](https://mlsanity.com/paper/1802.00923/) (2018)
- [Words Can Shift: Dynamically Adjusting Word Representations Using Nonverbal Behaviors](https://mlsanity.com/paper/1811.09362/) (2018)
- [GCNet: Graph Completion Network for Incomplete Multimodal Learning in Conversation](https://mlsanity.com/paper/2203.02177/) (2022)
- [Beneath the Tip of the Iceberg: Current Challenges and New Directions in Sentiment Analysis Research](https://mlsanity.com/paper/2005.00357/) (2020)

## Built on by (most cited first, 20 of 20)

- [MAE-DFER: Efficient Masked Autoencoder for Self-supervised Dynamic Facial Expression Recognition](https://mlsanity.com/paper/2307.02227/) (2023)
- [MER 2023: Multi-label Learning, Modality Robustness, and Semi-Supervised Learning](https://mlsanity.com/paper/2304.08981/) (2023)
- [HiCMAE: Hierarchical Contrastive Masked Autoencoder for Self-Supervised Audio-Visual Emotion Recognition](https://mlsanity.com/paper/2401.05698/) (2024)
- [MER 2024: Semi-Supervised Learning, Noise Robustness, and Open-Vocabulary Multimodal Emotion Recognition](https://mlsanity.com/paper/2404.17113/) (2024)
- [SDR-GNN: Spectral Domain Reconstruction Graph Neural Network for Incomplete Multimodal Learning in Conversational Emotion Recognition](https://mlsanity.com/paper/2411.19822/) (2024)
- [Modality-Invariant Bidirectional Temporal Representation Distillation Network for Missing Multimodal Sentiment Analysis](https://mlsanity.com/paper/2501.05474/) (2025)
- [Towards Robust Multimodal Emotion Recognition under Missing Modalities and Distribution Shifts](https://mlsanity.com/paper/2506.10452/) (2025)
- [DUAL: Dynamic Uncertainty-Aware Learning](https://mlsanity.com/paper/2506.03158/) (2025)
- [Towards Robust and Realible Multimodal Misinformation Recognition with Incomplete Modality](https://mlsanity.com/paper/2510.05839/) (2025)
- [Calibrating Multimodal Consensus for Emotion Recognition](https://mlsanity.com/paper/2510.20256/) (2025)
- [DCER: Dual-Stage Compression and Energy-Based Reconstruction](https://mlsanity.com/paper/2602.04904/) (2026)
- [CorMulT: A Semi-supervised Modality Correlation-aware Multimodal Transformer for Sentiment Analysis](https://mlsanity.com/paper/2407.07046/) (2024)
- [Robust Multimodal Sentiment Analysis via Double Information Bottleneck](https://mlsanity.com/paper/2511.01444/) (2025)
- [DashFusion: Dual-stream Alignment with Hierarchical Bottleneck Fusion for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2512.05515/) (2025)
- [Hierarchical Adaptive Expert for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2503.22715/) (2025)
- [QiNN-QJ: A Quantum-inspired Neural Network with Quantum Jump for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2510.27091/) (2025)
- [OmniMER: Auxiliary-Enhanced LLM Adaptation for Indonesian Multimodal Emotion Recognition](https://mlsanity.com/paper/2512.19379/) (2025)
- [MRUF: Multi-granularity Routing with Uncertainty-Aware Fusion for Robust Multimodal Sentiment Analysis](https://mlsanity.com/paper/2607.10599/) (2026)
- [MIDAS: Mutual Information Disentanglement with Uncertainty-Aware Fusion for Incomplete Multimodal Sentiment Analysis](https://mlsanity.com/paper/2608.09986/) (2026)
- [SemMSA: Latent Semantic-Aided Robust Multimodal Sentiment Analysis with Incomplete Data](https://mlsanity.com/paper/2609.30238/) (2026)

## Abstract (arXiv)

With the proliferation of user-generated online videos, Multimodal Sentiment Analysis (MSA) has attracted increasing attention recently. Despite significant progress, there are still two major challenges on the way towards robust MSA: 1) inefficiency when modeling cross-modal interactions in unaligned multimodal data; and 2) vulnerability to random modality feature missing which typically occurs in realistic settings. In this paper, we propose a generic and unified framework to address them, named Efficient Multimodal Transformer with Dual-Level Feature Restoration (EMT-DLFR). Concretely, EMT employs utterance-level representations from each modality as the global multimodal context to interact with local unimodal features and mutually promote each other. It not only avoids the quadratic scaling cost of previous local-local cross-modal interaction methods but also leads to better performance. To improve model robustness in the incomplete modality setting, on the one hand, DLFR performs low-level feature reconstruction to implicitly encourage the model to learn semantic information from incomplete data. On the other hand, it innovatively regards complete and incomplete data as two different views of one sample and utilizes siamese representation learning to explicitly attract their high-level representations. Comprehensive experiments on three popular datasets demonstrate that our method achieves superior performance in both complete and incomplete modality settings.