Integrating Contrastive Learning into a Multitask Transformer Model for Effective Domain Adaptation

2023-10-07 · Chung-Soo Ahn, Jagath C. Rajapakse, Rajib Rana

While speech emotion recognition (SER) research has made significant progress, achieving generalization across various corpora continues to pose a problem. We propose a novel domain adaptation technique that embodies a multitask framework with SER as the primary task, and contrastive learning and information maximisation loss as auxiliary tasks, underpinned by fine-tuning of transformers pre-trained on large language models. Empirical results obtained through experiments on well-established datasets like IEMOCAP and MSP-IMPROV, illustrate that our proposed model achieves state-of-the-art performance in SER within cross-corpus scenarios.

📄 PDF Abstract BibTeX arXiv:2310.04703

Code (0)

등록된 구현이 없습니다.

Tasks

Contrastive LearningCross-corpusDomain Adaptation Emotion Recognition Speech Emotion Recognition

Methods 이 논문이 사용한 방법론

Contrastive Learning 설명 없음

Contrastive and Multi-Task Learning on Noisy Brain Signals with Nonlinear Dynamical Signatures

2026-01-13 · Sucheta Ghosh, Felix Dietrich, Zahra Monfared arxiv

We introduce a two-stage multitask learning framework for analyzing Electroencephalography (EEG) signals that integrates denoising, dynamical modeling, and representation learning. In the first stage, a denoising autoenc…

Self-Supervised Learning Representation Learning Multi-Task Learning Eeg Decoding

MulT: An End-to-End Multitask Learning Transformer

2022-05-17 · CVPR 2022 1 · Deblina Bhattacharjee, Tong Zhang, Sabine Süsstrunk, Mathieu Salzmann

We propose an end-to-end Multitask Learning Transformer framework, named MulT, to simultaneously learn multiple high-level vision tasks, including depth estimation, semantic segmentation, reshading, surface normal estima…

Decoder Depth Estimation Edge Detection Keypoint Detection+2

Vision Transformer Adapters for Generalizable Multitask Learning

2023-08-23 · ICCV 2023 1 · Deblina Bhattacharjee, Sabine Süsstrunk, Mathieu Salzmann

We introduce the first multitasking vision transformer adapters that learn generalizable task affinities which can be applied to novel tasks and domains. Integrated into an off-the-shelf vision transformer backbone, our …

Domain Adaptation Unsupervised Domain Adaptation

OmniVec2 - A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning

2024-01-01 · CVPR 2024 1 · Siddharth Srivastava, Gaurav Sharma

We present a novel multimodal multitask network and associated training algorithm. The method is capable of ingesting data from approximately 12 different modalities namely image video audio text depth point cloud ti…

3D Point Cloud Classification Action Classification Action Recognition Audio Classification+6

OmniVec2 -- A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning

2025-07-06 · Siddharth Srivastava, Gaurav Sharma arxiv