Fengling Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Fengling Li has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
J. Lu
China University of Geosciences(Beijing)
1 shared publication- Jianjun Gao
School of Geophysics and Information Technology, China University of Geosciences, Beijing, China
1 shared publication - Jingjing Li
University of Electronic Science and Technology of China, Chengdu, China
1 shared publication - Lei Zhu
Tongji University, Shanghai, China
1 shared publication - Yang Wang
State Key Laboratory of Tropic Ocean Engineering Materials and Materials Evaluation, School of Materials Science and Engineering, Hainan University, Haikou, China.
1 shared publication - Yang Yang
University of Electronic Science and Technology of China, Chengdu, China
1 shared publication
Latest Publications
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 2654-2662
Noise-Robust Continual Test-Time Domain Adaptation
Continual test-time domain adaptation (TTA) is a challenging topic in the field of source-free domain adaptation, which focuses on addressing cross-domain multimedia information during inference with a continuously changing data distribution. Previous methods have been found to lack noise robustness, leading to a significant increase in errors under strong noise. In this paper, we address the noise-robustness problem in continual TTA by offering three effective recipes to mitigate it. At the category level, we employ the Taylor cross-entropy loss to alleviate the low confidence category bias commonly associated with cross-entropy. At the sample level, we reweight the target samples based on uncertainty to prevent the model from overfitting on noisy samples. Finally, to reduce pseudo-label noise, we propose a soft ensemble negative learning mechanism to guide the model optimization using ensemble complementary pseudo labels. Our method achieves state-of-the-art performance on three widely used continual TTA datasets, particularly in the strong noise setting that we introduced.
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 2654-2662
Open AccessNoise-Robust Continual Test-Time Domain Adaptation
Continual test-time domain adaptation (TTA) is a challenging topic in the field of source-free domain adaptation, which focuses on addressing cross-domain multimedia information during inference with a continuously changing data distribution. Previous methods have been found to lack noise robustness, leading to a significant increase in errors under strong noise. In this paper, we address the noise-robustness problem in continual TTA by offering three effective recipes to mitigate it. At the category level, we employ the Taylor cross-entropy loss to alleviate the low confidence category bias commonly associated with cross-entropy. At the sample level, we reweight the target samples based on uncertainty to prevent the model from overfitting on noisy samples. Finally, to reduce pseudo-label noise, we propose a soft ensemble negative learning mechanism to guide the model optimization using ensemble complementary pseudo labels. Our method achieves state-of-the-art performance on three widely used continual TTA datasets, particularly in the strong noise setting that we introduced.
- 1 January 2021
- Published by EAGE Publications bv
- Vol. 2021 (1) , 1-5
Multi-component seismic data reconstruction based on vector POCS method via complexified quaternion fourier transform
- 1 January 2021
- Published by EAGE Publications bv
- Vol. 2021 (1) , 1-5
Multi-component seismic data reconstruction based on vector POCS method via complexified quaternion fourier transform