paper-with-me

홈 › Papers

Cross-Task Consistency Learning Framework for Multi-Task Learning

2021-11-28 · Akihiro Nakano, Shi Chen, Kazuyuki Demachi

Multi-task learning (MTL) is an active field in deep learning in which we train a model to jointly learn multiple tasks by exploiting relationships between the tasks. It has been shown that MTL helps the model share the learned features between tasks and enhance predictions compared to when learning each task independently. We propose a new learning framework for 2-task MTL problem that uses the predictions of one task as inputs to another network to predict the other task. We define two new loss terms inspired by cycle-consistency loss and contrastive learning, alignment loss and cross-task consistency loss. Both losses are designed to enforce the model to align the predictions of multiple tasks so that the model predicts consistently. We theoretically prove that both losses help the model learn more efficiently and that cross-task consistency loss is better in terms of alignment with the straight-forward predictions. Experimental results also show that our proposed model achieves significant performance on the benchmark Cityscapes and NYU dataset.

📄 PDF Abstract BibTeX arXiv:2111.14122

Code (1)

akarimoon/xtask_mt 공식 구현 pytorch

Tasks

Contrastive LearningMulti-Task Learning

Similar Papers 제목 키워드 기반

Robust Learning Through Cross-Task Consistency

2020-06-07 · Amir Zamir, Alexander Sax, Teresa Yeo, Oğuzhan Kar 외

Visual perception entails solving a wide set of tasks, e.g., object detection, depth estimation, etc. The predictions made for multiple tasks from the same image are not independent, and therefore, are expected to be con…

3D ReconstructionDepth EstimationMulti-Task Learningobject-detection+2

Robust Learning Through Cross-Task Consistency

2020-06-01 · CVPR 2020 6 · Amir R. Zamir, Alexander Sax, Nikhil Cheerla, Rohan Suri 외

Visual perception entails solving a wide set of tasks (e.g., object detection, depth estimation, etc). The predictions made for different tasks out of one image are not independent, and therefore, are expected to be 'con…

3D ReconstructionDepth EstimationMonocular Depth Estimationobject-detection+3

HCDG: A Hierarchical Consistency Framework for Domain Generalization on Medical Image Segmentation

2021-09-13 · Yijun Yang, Shujun Wang, Lei Zhu, Lequan Yu

Modern deep neural networks struggle to transfer knowledge and generalize across diverse domains when deployed to real-world applications. Currently, domain generalization (DG) is introduced to learn a universal represen…

Data AugmentationDomain GeneralizationImage SegmentationMedical Image Segmentation+3

Beyond Accuracy: Benchmarking Cross-Task Consistency in Unified Multimodal Models

2026-04-27 · Weixing Wang, Liudvikas Zekas, Anton Hackl, Constantin Alexander Auga 외 arxiv

Unified Multimodal Models (uMMs) aim to support both visual understanding and visual generation within a shared representation. However, existing evaluation protocols assess these two capabilities independently and do no…

Contrastive Multi-Task Dense Prediction

2023-07-16 · Siwei Yang, Hanrong Ye, Dan Xu

This paper targets the problem of multi-task dense prediction which aims to achieve simultaneous learning and inference on a bunch of multiple dense prediction tasks in a single framework. A core objective in design is h…

Contrastive LearningPredictionRepresentation Learning