no code implementations • COLING 2022 • BoWen Zhang, Xu Huang, Zhichao Huang, Hu Huang, Baoquan Zhang, Xianghua Fu, Liwen Jing
SILTN is interpretable because it is a neurosymbolic formalism and a computational model that supports learning and reasoning about data with a differentiable first-order logic language (FOL).
1 code implementation • 26 Apr 2024 • Jiajun Liang, Baoquan Zhang, Yunming Ye, Xutao Li, Chuyao Luo, Xukai Fu
Different from the previous models, MCSDNet targets on multi-frames detection and leverages multi-scale spatiotemporal information for the detection of MCS regions in remote sensing imagery(RSI).
1 code implementation • 17 Mar 2024 • Fuqiang Niu, Min Yang, Ang Li, Baoquan Zhang, Xiaojiang Peng, BoWen Zhang
Previous stance detection studies typically concentrate on evaluating stances within individual instances, thereby exhibiting limitations in effectively modeling multi-party discussions concerning the same specific topic, as naturally transpire in authentic social media interactions.
no code implementations • 15 Mar 2024 • Baoquan Zhang, Huaibin Wang, Luo Chuyao, Xutao Li, Liang Guotao, Yunming Ye, Xiaochen Qi, Yao He
To this end, we propose a novel codebook transfer framework with part-of-speech, called VQCT, which aims to transfer a well-trained codebook from pretrained language models to VQIM for robust codebook learning.
1 code implementation • 11 Dec 2023 • Demin Yu, Xutao Li, Yunming Ye, Baoquan Zhang, Chuyao Luo, Kuai Dai, Rui Wang, Xunlai Chen
A unified and flexible framework that can equip any type of spatio-temporal models is proposed based on residual diffusion, which effectively tackles the shortcomings of previous methods.
1 code implementation • 26 Sep 2023 • Huiwei Lin, Shanshan Feng, Baoquan Zhang, Xutao Li, Yew-Soon Ong, Yunming Ye
Inspired by this finding, we propose a novel replay-based method called proxy-based contrastive replay (PCR), which replaces anchor-to-sample pairs with anchor-to-proxy pairs in the contrastive-based loss to alleviate the phenomenon of forgetting.
no code implementations • 8 Sep 2023 • Huiwei Lin, Shanshan Feng, Baoquan Zhang, Hongliang Qiao, Xutao Li, Yunming Ye
By decomposing the dot-product logits into an angle factor and a norm factor, we empirically find that the bias problem mainly occurs in the angle factor, which can be used to learn novel knowledge as cosine logits.
no code implementations • 31 Jul 2023 • Baoquan Zhang, Chuyao Luo, Demin Yu, Huiwei Lin, Xutao Li, Yunming Ye, BoWen Zhang
Its key idea is learning a deep model in a bi-level optimization manner, where the outer-loop process learns a shared gradient descent algorithm (i. e., its hyperparameters), while the inner-loop process leverage it to optimize a task-specific model by using only few labeled data.
1 code implementation • CVPR 2023 • Huiwei Lin, Baoquan Zhang, Shanshan Feng, Xutao Li, Yunming Ye
It aims to continuously learn new classes from data stream and the samples of data stream are seen only once, which suffers from the catastrophic forgetting issue, i. e., forgetting historical knowledge of old classes.
no code implementations • 3 Mar 2022 • Baoquan Zhang, Hao Jiang, Xutao Li, Shanshan Feng, Yunming Ye, Rui Ye
Then, resorting to the prior, we split each few-shot task to a set of subtasks with different concept levels and then perform class prediction via a model of decision tree.
no code implementations • 9 Oct 2021 • Baoquan Zhang, Shanshan Feng, Xutao Li, Yunming Ye, Rui Ye
In this framework, a scene graph construction module is carefully designed to represent each test remote sensing image or each scene class as a scene graph, where the nodes reflect these co-occurrence objects meanwhile the edges capture the spatial correlations between these co-occurrence objects.
1 code implementation • 11 Aug 2021 • Baoquan Zhang, Xutao Li, Yunming Ye, Shanshan Feng
In this paper, 1) we figure out the reason, i. e., in the pre-trained feature space, the base classes already form compact clusters while novel classes spread as groups with large variances, which implies that fine-tuning feature extractor is less meaningful; 2) instead of fine-tuning feature extractor, we focus on estimating more representative prototypes.
1 code implementation • 26 Mar 2021 • Baoquan Zhang, Xutao Li, Shanshan Feng, Yunming Ye, Rui Ye
Although the existing meta-optimizers can also be adapted to our framework, they all overlook a crucial gradient bias issue, \emph{i. e.}, the mean-based gradient estimation is also biased on sparse data.
1 code implementation • CVPR 2021 • Baoquan Zhang, Xutao Li, Yunming Ye, Zhichao Huang, Lisai Zhang
To avoid the prototype completion error caused by primitive knowledge noises or class differences, we further develop a Gaussian based prototype fusion strategy that combines the mean-based and completed prototypes by exploiting the unlabeled samples.
no code implementations • 5 Jul 2020 • Baoquan Zhang, Ka-Cheong Leung, Yunming Ye, Xutao Li
To this end, we propose a novel meta-learning framework, called MetaConcept, which learns to abstract concepts via the concept graph.