Bi-Bimodal Modality Fusion for Correlation-Controlled Multimodal Sentiment Analysis
Open accessThe DOI matches its Crossref record; 18 of the 18 references it cites were found in a registry or an index.
Abstract
Multimodal sentiment analysis aims to extract and integrate semantic information collected from multiple modalities to recognize the expressed emotions and sentiment in multimodal data. This research area’s major concern lies in developing an extraordinary fusion scheme that can extract and integrate key information from various modalities. However, previous work is restricted by the lack of leveraging dynamics of independence and correlation between modalities to reach top performance. To mitigate this, we propose the Bi-Bimodal Fusion Network (BBFN), a novel end-to-end network that performs fusion (relevance increment) and separation (difference increment) on pairwise modality representations. The two parts are trained simultaneously such that the combat between them is simulated. The model takes two bimodal pairs as input due to the known information imbalance among modalities. In addition, we leverage a gated control mechanism in the Transformer architecture to further improve the final output. Experimental results on three datasets (CMU-MOSI, CMU-MOSEI, and UR-FUNNY) verifies that our model significantly outperforms the SOTA. The implementation of this work is available at https://github.com/declare-lab/multimodal-deep-learning and https://github.com/declare-lab/BBFN.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., et al. (2020) · arXiv · doi:10.48550/arxiv.2010.11929 · Cited by 47,472
Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1412.3555 · Cited by 11,008
Graves, A., Fernández, S., Gomez, F., & Schmidhuber, J. (2006) · Proceedings of the 23rd international conference on Machine learning - ICML '06 · doi:10.1145/1143844.1143891 · Cited by 5,370
Multimodal Transformer for Unaligned Multimodal Language Sequences
Tsai, Y.-H. H., Bai, S., Liang, P. P., Kolter, J. Z., Morency, L.-P., & Salakhutdinov, R. (2019) · Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/p19-1656 · Cited by 2,259 · cited 6×
“Since it has been empirically shown that the text modality is most significant [26, 34], our model takes two text-related modality pairs, TV (text-visual) and TA (textacoustic), as respective inputs for its two bimodal learning modules.”
BackgroundMethodInfluentialSelf-citationTensor Fusion Network for Multimodal Sentiment Analysis
Zadeh, A., Chen, M., Poria, S., Cambria, E., & Morency, L.-P. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1115 · Cited by 2,102 · cited 4×
“• Tensor FusionNetwork (TFN) [43]: This approachmodels intra-modality and inter-modality dynamics concurrently with local feature extraction network and 3-fold Cartesian product.”
BackgroundInfluentialSelf-citationMultimodal Language Analysis in the Wild: CMU-MOSEI Dataset and Interpretable Dynamic Fusion Graph
Bagher Zadeh, A., Liang, P. P., Poria, S., Cambria, E., & Morency, L.-P. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p18-1208 · Cited by 1,433 · cited 1×
“The CMU-MOSEI dataset [46] is an upgraded version of CMU-MOSI concerning the number of samples.”
MethodSelf-citationMultimodal fusion for multimedia analysis: a survey
Atrey, P. K., Hossain, M. A., El Saddik, A., & Kankanhalli, M. S. (2010) · Multimedia Systems · doi:10.1007/s00530-010-0182-0 · Cited by 1,279
Efficient Low-rank Multimodal Fusion With Modality-Specific Factors
Liu, Z., Shen, Y., Lakshminarasimhan, V. B., Liang, P. P., Bagher Zadeh, A., & Morency, L.-P. (2018) · Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p18-1209 · Cited by 1,170 · cited 2×
“• Low-rank Matrix Fusion (LMF) [18]: It decomposes highorder tensors into many low-rank factors then performs efficient fusion based on these factors.”
BackgroundSelf-citationMISA: Modality-Invariant and -Specific Representations for Multimodal Sentiment Analysis
Hazarika, D., Zimmermann, R., & Poria, S. (2020) · Proceedings of the 28th ACM International Conference on Multimedia · doi:10.1145/3394171.3413678 · Cited by 1,073 · cited 8×
“To better illustrate how BBFN beats the SOTA (MISA), we compute the absolute errors of all the predictions on the test set of CMU-MOSEI and paint their distributions in Fig.”
BackgroundMethodInfluentialSelf-citationMultimodal Sentiment Intensity Analysis in Videos: Facial Gestures and Verbal Messages
Zadeh, A., Zellers, R., Pincus, E., & Morency, L.-P. (2016) · IEEE Intelligent Systems · doi:10.1109/mis.2016.94 · Cited by 852
Self-citationPham, H., Liang, P. P., Manzini, T., Morency, L.-P., & Póczos, B. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.33016892 · Cited by 512
Self-citationTowards multimodal sentiment analysis: harvesting opinions from the web
Morency, L.-P., Mihalcea, R., & Doshi, P. (2011) · Proceedings of the 13th international conference on multimodal interfaces · doi:10.1145/2070481.2070509 · Cited by 470
Self-citationSun, Z., Sarma, P., Sethares, W., & Liang, Y. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i05.6431 · Cited by 410
Youtube movie reviews: Sentiment analysis in an audio-visual context
Wollmer, M., Weninger, F., Knaup, T., Schuller, B., Sun, C., Sagae, K., & Morency, L.-P. (2013) · IEEE Intelligent Systems · doi:10.1109/mis.2013.34 · Cited by 403
Self-citationMultimodal sentiment analysis with word-level fusion and reinforcement learning
Chen, M., Wang, S., Liang, P. P., Baltrušaitis, T., Zadeh, A., & Morency, L.-P. (2017) · Proceedings of the 19th ACM International Conference on Multimodal Interaction · doi:10.1145/3136755.3136801 · Cited by 334 · cited 2×
“However, as it has been pointed out by several past research [6, 32, 34], the task-related information is not evenly distributed between the modalities.”
BackgroundMethodSelf-citationUR-FUNNY: A Multimodal Language Dataset for Understanding Humor
Hasan, M. K., Rahman, W., Bagher Zadeh, A., Zhong, J., Tanveer, M. I., Morency, L.-P., & Hoque, M. (. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1211 · Cited by 174 · cited 5×
“For experiments on UR-FUNNY we use another facial behavioral analysis tool, Openface [3] to capture facial gesture variance of Acoustic Modality.”
BackgroundInfluentialSelf-citationMultimodal Transformer for Multimodal Machine Translation
Yao, S., & Wan, X. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.400 · Cited by 152 · cited 1×
“Besides multimodal sentiment analysis, multimodal learning has been applied in many other language tasks, such as Machine Translation (MT) [12, 17, 30, 40], Named Entity Recognition (NER) [19, 21, 41, 47], and parsing [28, 29, 48].”
Method
Cited by
Gandhi, A., Adhvaryu, K., Poria, S., Cambria, E., & Hussain, A. (2023) · Information Fusion · doi:10.1016/j.inffus.2022.09.025 · Cited by 783
“[38] BBFN 2021 Bimodal Fusion CMU-MOSEI An innovative end-to-end Bimodal Fusion network that conducts fusion (relevance increment) and separation (difference increment) on pairs modality representations is introduced.”
BackgroundSelf-citationMulti-modal cross-attention network for Alzheimer’s disease diagnosis with multi-modality data
Zhang, J., He, X., Liu, Y., Cai, Q., Chen, H., & Qing, L. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.107050 · Cited by 117
Self-citationLiu, F., Shen, S.-Y., Fu, Z.-W., Wang, H.-Y., Zhou, A.-M., & Qi, J.-Y. (2022) · Entropy · doi:10.3390/e24071010 · Cited by 21
Bimodal fusion network with multi-head attention for multimodal sentiment analysis
Zhang, R., Xue, C., Qi, Q., Lin, L., Zhang, J., & Zhang, L. (2023) · Applied Sciences · doi:10.3390/app13031915 · Cited by 11
Unbiased Missing-Modality Multimodal Learning
Dai, R., Li, C., Yan, Y., Mo, L., Qin, K., & He, T. (2025) · 2025 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02272 · Cited by 11
Metadata
- Authors
- Han, W., Chen, H., Gelbukh, A., Zadeh, A., Morency, L.-P., & Poria, S.
- Venue
- Proceedings of the 2021 International Conference on Multimodal Interaction
- Publisher
- ACM
- Published
- 2021-10-18
- Pages
- 6-15
- OpenAlex
- W3183580551
- Cited by
- 232
- References
- 18
- FWCI
- 17.65
- Type
- conference paper
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.