mlsanityData to 14 Oct 2026

DLF: Disentangled-Language-Focused Multimodal Sentiment Analysis

arXiv 2412.12225 ↗First posted 16 Dec 2024cs.LG cs.AI cs.CLcited by 27

Not on the map. Why?

Abstract (arXiv)

Multimodal Sentiment Analysis (MSA) leverages heterogeneous modalities, such as language, vision, and audio, to enhance the understanding of human sentiment. While existing models often focus on extracting shared information across modalities or directly fusing heterogeneous modalities, such approaches can introduce redundancy and conflicts due to equal treatment of all modalities and the mutual transfer of information between modality pairs. To address these issues, we propose a Disentangled-Language-Focused (DLF) multimodal representation learning framework, which incorporates a feature disentanglement module to separate modality-shared and modality-specific information. To further reduce redundancy and enhance language-targeted features, four geometric measures are introduced to refine the disentanglement process. A Language-Focused Attractor (LFA) is further developed to strengthen language representation by leveraging complementary modality-specific information through a language-guided cross-attention mechanism. The framework also employs hierarchical predictions to improve overall accuracy. Extensive experiments on two popular MSA datasets, CMU-MOSI and CMU-MOSEI, demonstrate the significant performance gains achieved by the proposed DLF framework. Comprehensive ablation studies further validate the effectiveness of the feature disentanglement module, language-focused attractor, and hierarchical predictions. Our code is available at https://github.com/pwang322/DLF.

Key facts. Cited by 27 papers in the map (data through 14 Oct 2026). Builds on 16 of its 42 references that are in the map. First posted on arXiv 16 Dec 2024.

Builds on

16 of 42 references.

References · 16

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2015FaceNet: A Unified Embedding for Face Recognition and Clusteringcited by 2,849
  3. 2019Multimodal Transformer for Unaligned Multimodal Language Sequencescited by 645
  4. 2023NExT-GPT: Any-to-Any Multimodal LLMcited by 615
  5. 2017Tensor Fusion Network for Multimodal Sentiment Analysiscited by 433
  6. 2022Multimodal Learning with Transformers: A Surveycited by 400
  7. 2018Efficient Low-rank Multimodal Fusion with Modality-Specific Factorscited by 270
  8. 2020MISA: Modality-Invariant and -Specific Representations for Multimodal Sentiment Analysiscited by 265
  9. 2021Learning Modality-Specific Representations with Self-Supervised Multi-Task Learning for Multimodal Sentiment Analysiscited by 180
  10. 2015Cross Modal Distillation for Supervision Transfercited by 175
  11. 2018Learning Factorized Multimodal Representationscited by 171
  12. 2018Memory Fusion Network for Multi-view Sequential Learningcited by 168
4 more
  1. 2018Found in Translation: Learning Robust Joint Representations by Cyclic Translations Between Modalitiescited by 149
  2. 2024Correlation-Decoupled Knowledge Distillation for Multimodal Sentiment Analysis with Incomplete Modalitiescited by 32
  3. 2024WisdoM: Improving Multimodal Sentiment Analysis by Fusing Contextual World Knowledgecited by 16
  4. 2023Text-oriented Modality Reinforcement Network for Multimodal Sentiment Analysis from Unaligned Multimodal Sequencescited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 7; 2026: 20.

27 papers.

Sort:

By area:

  1. 2025DecAlign: Hierarchical Cross-Modal Alignment for Decoupled Multimodal Representation Learningcited by 24
  2. 2026Missing-by-Design: Certifiable Modality Deletion for Revocable Multimodal Sentiment Analysiscited by 6
  3. 2026Tri-Subspaces Disentanglement for Multimodal Sentiment Analysiscited by 6
  4. 2026Motion-Aware Caching for Efficient Autoregressive Video Generationcited by 6
  5. 2026CLCR: Cross-Level Semantic Collaborative Representation for Multimodal Learningcited by 5
  6. 2025Towards Robust Multimodal Emotion Recognition under Missing Modalities and Distribution Shiftscited by 4
  7. 2026Temporal-Spatial Decouple before Act: Disentangled Representation Learning for Multimodal Sentiment Analysiscited by 3
  8. 2026Group Cognition Learning: Making Everything Better Through Governed Two-Stage Agents Collaborationcited by 3
  9. 2025Disentangling Bias by Modeling Intra- and Inter-modal Causal Attention for Multimodal Sentiment Analysiscited by 2
  10. 2025Towards Minimal Causal Representations for Human Multimodal Language Understandingcited by 2
  11. 2025Calibrating Multimodal Consensus for Emotion Recognitioncited by 2
  12. 2026Emotion Collider: Dual Hyperbolic Mirror Manifolds for Sentiment Recovery via Anti Emotion Reflectioncited by 2
15 more
  1. 2026CaReFlow: Cyclic Adaptive Rectified Flow for Multimodal Fusioncited by 2
  2. 2025Uncertainty-Aware Collaborative System of Large and Small Models for Multimodal Sentiment Analysiscited by 1
  3. 2026Decoupled Hierarchical Distillation for Multimodal Emotion Recognitioncited by 1
  4. 2026ModalImmune: Immunity Driven Unlearning via Self Destructive Trainingcited by 1
  5. 2026Addressing Missing and Noisy Modalities in One Solution: Unified Modality-Quality Framework for Low-quality Multimodal Datacited by 1
  6. 2026Progressive Representation Learning for Multimodal Sentiment Analysis with Incomplete Modalitiescited by 1
  7. 2026Enhance-then-Balance Modality Collaboration for Robust Multimodal Sentiment Analysiscited by 1
  8. 2026Controlling Decision Drift in Multimodal Sentiment Analysis with Missing Modalitiescited by 1
  9. 2025QiNN-QJ: A Quantum-inspired Neural Network with Quantum Jump for Multimodal Sentiment Analysiscited by 0
  10. 2026GRCF: Two-Stage Groupwise Ranking and Calibration Framework for Multimodal Sentiment Analysiscited by 0
  11. 2026Breaking the Correlation Plateau: On the Optimization and Capacity Limits of Attention-Based Regressorscited by 0
  12. 2026Segregate, Refine, Integrate: Decomposing Multimodal Fusion for Sentiment Analysiscited by 0
  13. 2026Vision-Guided Text Prompt Tuning for Multimodal Sentiment Analysiscited by 0
  14. 2026SemMSA: Latent Semantic-Aided Robust Multimodal Sentiment Analysis with Incomplete Datacited by 0
  15. 2026Reliability-aware Cross-sample Enhancement for Robust Multimodal Sentiment Analysiscited by 0