Kun Wang

Kun Wang

Auto-generated profile, based on publicly available information

Kun Wang

Nanyang technological university, Singapore, Singapore

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 August 2025
Verified
Conference Paper
Open access
MYMiao Yu
Miao Yu
FMFanci Meng
Fanci Meng
XZXinyun Zhou
Xinyun Zhou
SWShilong Wang
Shilong Wang
JMJunyuan Mao
Junyuan Mao
QWQingsong Wen
Qingsong Wen

1 January 2025
Verified
Conference Paper
Open access
PDF file
YYYanwei Yue
Yanwei Yue
GZGuibin Zhang
Guibin Zhang
BLBoyang Liu
Boyang Liu
GWGuancheng Wan
Guancheng Wan
KWKun Wang
Kun Wang
YQYiyan Qi
Yiyan Qi
Yanwei Yue, Guibin Zhang, Boyang Liu, Guancheng Wan, Kun Wang, Dawei Cheng, Yiyan Qi. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

4 August 2023
Verified
Conference Paper
Open access
ZZZhengyang Zhou
Zhengyang Zhou
QHQihe Huang
Qihe Huang
KYKuo Yang
Kuo Yang
KWKun Wang
Kun Wang
XWXu Wang
Xu Wang
YWYang Wang
Yang Wang
Spatiotemporal (ST) learning has become a crucial technique for urban digitalization. Due to expansions and dynamics of cities, current spatiotemporal models are inclined to suffer distribution shifts between training and testing sets, leading to the OOD delimma. However, few studies focus on such OOD problem in temporal regressions, let alone spatiotemporal learning. Spatiotemporal data usually reveals segment-level heterogeneity within periodicity and complex spatial dependencies, posing challenges to invariance extraction. In this paper, we find that ST relations make sense for generalization and devise a Causal ST learning framework, CauSTG, which enables invariant relation transferred to OOD scenarios. Specifically, we take temporal steps as environments, and transform spatial-temporal relations into learnable parameters. To tackle heterogeneity in periodicity, we partition temporal steps into sub-environments by identifying distinctive trend patterns, enabling re-organized samples trained separately. To extract invariance within ST observations, we propose a spatiotemporal consistency learner and a hierarchical invariance explorer to jointly filter out stable relations. Our spatiotemporal learner quantifies bi-directional spatial consistency and extracts disentangled seasonal-trend patterns via trainable parameters. Further, the hierarchical invariance explorer constructs variation-based filter to achieve both local and global invariances. Experiments reveal that CauSTG can increase at most 10.26% performance against best baselines, and visualized invariant relations can well interpret the physical rationales. The appendix and codes can be available in our Github repository.

13 December 2023 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
KWKun Wang
Kun Wang
YLYuxuan Liang
Yuxuan Liang
XLXinglin Li
Xinglin Li
GLGuohao Li
Guohao Li
BGBernard Ghanem
Bernard Ghanem
YWYang Wang
Yang Wang
The training and inference of Graph Neural Networks (GNNs) are costly when scaling up to large-scale graphs. Graph Lottery Ticket (GLT) has presented the first attempt to accelerate GNN inference on large-scale graphs by jointly pruning the graph structure and the model weights. Though promising, GLT encounters robustness and generalization issues when deployed in real-world scenarios, which are also long-standing and critical problems in deep learning ideology. In real-world scenarios, the distribution of unseen test data is typically diverse. We attribute the failures on out-of-distribution (OOD) data to the incapability of discerning causal patterns, which remain stable amidst distribution shifts. In traditional spase graph learning, the model performance deteriorates dramatically as the graph/network sparsity exceeds a certain high level. Worse still, the pruned GNNs are hard to generalize to unseen graph data due to limited training set at hand. To tackle these issues, we propose the Resilient Graph Lottery Ticket (RGLT) to find more robust and generalizable GLT in GNNs. Concretely, we reactivate a fraction of weights/edges by instantaneous gradient information at each pruning point. After sufficient pruning, we conduct environmental interventions to extrapolate potential test distribution. Finally, we perform last several rounds of model averages to further improve generalization. We provide multiple examples and theoretical analyses that underpin the universality and reliability of our proposal. Further, RGLT has been experimentally verified across various independent identically distributed (IID) and out-of-distribution (OOD) graph benchmarks. The source code for this work is available at https://github.com/Lyccl/RGLT for PyTorch implementation.

13 May 2024
Verified
Conference Paper
Open access
JFJunfeng Fang
Junfeng Fang
XLXinglin Li
Xinglin Li
YSYongduo Sui
Yongduo Sui
YGYuan Gao
Yuan Gao
GZGuibin Zhang
Guibin Zhang
XHXiangnan He
Xiangnan He
Graph representation learning on vast datasets, like web data, has made significant strides. However, the associated computational and storage overheads raise concerns. In sight of this, Graph condensation (GCond) has been introduced to distill these large real datasets into a more concise yet information-rich synthetic graph. Despite acceleration efforts, existing GCond methods mainly grapple with efficiency, especially on expansive web data graphs. Hence, in this work, we pinpoint two major inefficiencies of current paradigms: (1) the concurrent updating of a vast parameter set, and (2) pronounced parameter redundancy. To counteract these two limitations correspondingly, we first (1) employ the Mean-Field variational approximation for convergence acceleration, and then (2) propose the objective of Gradient Information Bottleneck (GDIB) to prune redundancy. By incorporating the leading explanation techniques (e.g., GNNExplainer and GSAT) to instantiate the GDIB, our EXGC, the Efficient and eXplainable Graph Condensation method is proposed, which can markedly boost efficiency and inject explainability. Our extensive evaluations across eight datasets underscore EXGC's superiority and relevance. Code is available at https://github.com/MangoKiller/EXGC.

19 December 2025 in IEEE Signal Processing Magazine
Verified
Journal Article
Closed access
XZXINLIANG ZHOU
XINLIANG ZHOU
CLChenyu Liu
Chenyu Liu
ZCZhisheng Chen
Zhisheng Chen
KWKun Wang
Kun Wang
YDYi Ding
Yi Ding
QWQingsong Wen
Qingsong Wen
Brain foundation models (BFMs) represent a transformative paradigm in computational neuroscience that leverages large-scale pretraining on diverse neural signals to achieve robust generalization across tasks, modalities, and experimental contexts. This survey article establishes the first comprehensive definition and framework for BFMs, systematically examining their construction, core methodologies, and applications. We present key approaches for data processing and training strategies alongside diverse applications spanning brain decoding and scientific discovery. Through critical analysis of recent methodological innovations, we identify fundamental challenges that must be addressed to realize the full potential of BFMs, including advancing data quality and standardization, optimizing model architectures, improving training efficiency, and enhancing interpretability. By bridging the gap between neuroscience and artificial intelligence (AI), BFMs present unprecedented opportunities to revolutionize brain research, clinical diagnostics, and therapeutic interventions. This article serves as a foundational reference for researchers and practitioners seeking to understand and advance this emerging field.