Learning Robust Multi|Modal Representation for Multi|Label ...

Learning Robust Multi-Modal Representation for Multi-Label ...

This paper presents two adversarial training strategies to learn more robust multi-modal representation for multi-label emotion recognition.

Learning Robust Multi-Modal Representation for Multi-Label ...

Page 1. Learning Robust Multi-Modal Representation for Multi-Label. Emotion Recognition via Adversarial Masking and Perturbation. Shiping Ge.

Learning Robust Multi-Modal Representation for Multi-Label ...

Learning Robust Multi-Modal Representation for Multi-Label. Emotion Recognition via Adversarial Masking and Perturbation. Shiping Ge. Zhiwei ...

ShipingGe/MMER: Code for the paper Learning Robust Multi-Modal ...

Code for the paper Learning Robust Multi-Modal Representation for Multi-Label Emotion Recognition via Adversarial Masking and Perturbation - ShipingGe/MMER.

Quantifying and Enhancing Multi-modal Robustness with ... - arXiv

... representation margins and more ... Inspired by our theoretical finding, we introduce a training procedure called Certifiable Robust Multi ...

Multi-Label Multimodal Emotion Recognition With Transformer ...

Effectively incorporating information from multiple modalities in video data to learn robust multimodal representation for improving recognition ...

Robustness in Multimodal Learning under Train-Test Modality ...

We present a multimodal robustness framework to provide a systematic analysis of common multimodal representation learning methods.

[2403.06832] The Power of Noise: Toward a Unified Multi-modal ...

The advancement of Multi-modal Pre-training highlights the necessity for a robust Multi-Modal Knowledge Graph (MMKG) representation learning ...

Quantifying and Enhancing Multi-modal Robustness with Modality...

This study delves into the robustness of multi-modal models when faced with perturbations, including uni-modal attacks and missing modalities.

Adapt and explore: Multimodal mixup for representation learning

Improving the robustness against missing modalities and the performance of multimodal representations. •. Transferable to various multimodal learning scenarios.

LEARNING ROBUST REPRESENTATIONS VIA MULTI-VIEW ...

A theoretical analysis leads to the definition of a new multi-view model that produces state-of-the-art re- sults on the Sketchy dataset and label-limited ...

Robustness in Multimodal Learning under Train-Test Modality ...

AudioSet (Gemmeke et al., 2017) is a video, audio, and text multi-label audio classification dataset over 527 classes. Prior work has largely leveraged the ...

Robust Multi-View Representation Learning (Student Abstract)

Robust Multi-View Representation Learning (Student Abstract). Authors. Sibi Venkatesan Carnegie Mellon University; James K. Miller Carnegie Mellon University ...

GeWu-Lab/Certifiable-Robust-Multi-modal-Training - GitHub

Accepted by: The Twelfth International Conference on Learning Representations (ICLR 2024). [arXiv]. Multi-modal model is vulnerable toward a certain modality.

Adversarially Robust Multi-task Representation Learning

Poster. Adversarially Robust Multi-task Representation Learning. Austin Watkins · Thanh Nguyen-Tang · Enayat Ullah · Raman Arora. [ Abstract ].

Robust Multi-View Representation Learning

Robust Multi-View Representation Learning. Sibi Venkatesan, James K. Miller, and Artur Dubrawski. Conference Paper, Proceedings of 34th AAAI Conference on ...

Robust Multi-Modality Person Re-identification - AAAI

different modalities to unify the representations for images. Hao et al. (2019b) use sphere softmax to learn a hyper- sphere manifold embedding and constrain ...

Exploiting Multi-modal Fusion for Robust Face Representation ...

We propose a multi-modal fusion framework for robustly learning face representations in the presence of missing modalities, using a combination of RGB, depth, ...

Multi-View Robust Graph Representation Learning for Graph ... - IJCAI

We present MGRL, a multi-view representation learning model for graph classification tasks that achieves robust results.

Robust Multimodal Learning via Representation Decoupling

... Multimodal Representation Network (DMRNet) to assist robust multimodal learning. Specifically, DMRNet models the input from different modality ...