Author name cluster

Chuang Liu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

14 papers

2 author rows

AAAI Conference 2026 Conference Paper

From Subtle to Significant: Prompt-Driven Self-Improving Optimization in Test-Time Graph OOD Detection

Luzhi Wang
Xuanshuo Fu
He Zhang
Chuang Liu
Xiaobao Wang
Hongbo Liu

Graph Out-of-Distribution (OOD) detection aims to identify whether a test graph deviates from the distribution of graphs observed during training, which is critical for ensuring the reliability of Graph Neural Networks (GNNs) when deployed in open-world scenarios. Recent advances in graph OOD detection have focused on test-time training techniques that facilitate OOD detection without accessing potential supervisory information (e.g., training data). However, most of these methods employ a one-pass inference paradigm, which prevents them from progressively correcting erroneous predictions to amplify OOD signals. To this end, we propose a Self-Improving Graph Out-of-Distribution detector (SIGOOD), which is an unsupervised framework that integrates continuous self-learning with test-time training for effective graph OOD detection. Specifically, SIGOOD generates a prompt to construct a prompt-enhanced graph that amplifies potential OOD signals. To optimize prompts, SIGOOD introduces an Energy Preference Optimization (EPO) loss, which leverages energy variations between the original test graph and the prompt-enhanced graph. By iteratively optimizing the prompt by involving it into the detection model in a self-improving loop, the resulting optimal prompt-enhanced graph is ultimately used for OOD detection. Comprehensive evaluations on 21 real-world datasets confirm the effectiveness and outperformance of our SIGOOD method.

PDF Details DOI

EAAI Journal 2026 Journal Article

Structural-aware key node identification in hypergraphs via representation learning and fine-tuning

Xiaonan Ni
Guangyuan Mei
Su-Su Zhang
Yang Chen
Xin Xu
Chuang Liu
Xiu-Xiu Zhan

The ability to pinpoint strategically important nodes plays a decisive role in shaping diffusion outcomes and maintaining the stability of complex systems. Yet, most existing approaches remain rooted in pairwise interaction assumptions, making them ill-suited for systems where collective participation and attribute-sharing give rise to higher-order structures. In this work, we introduce AHGA, a learning-driven framework that leverages autoencoder-based representations, hypergraph neural network pre-training, and an active learning mechanism to uncover nodes that jointly influence propagation dynamics and structural cohesion. Rather than relying on handcrafted descriptors, AHGA learns informative higher-order features and progressively refines node importance through selective supervision. Evaluations on eight empirical hypergraphs show that this strategy leads to substantially more reliable rankings, with improvements of up to 36. 8% over classical baselines. Beyond ranking accuracy, nodes prioritized by AHGA exhibit pronounced structural leverage: their removal triggers an accelerated loss of network efficiency, reaching 0. 6628, markedly exceeding the disruptive effect achieved by competing methods. These findings demonstrate that AHGA not only advances higher-order node identification methodology, but also offers practical guidance for intervention strategies in scenarios such as misinformation containment and infrastructure robustness.

Details DOI

YNIMG Journal 2026 Journal Article

Treadmill exercise rebalances M1 response repertoires during motor state transitions in Parkinsonian mice

Chuang Liu
Yujie Xue
Chuanliang Han
Longwei Wei
Talifu Zikereya
Wei Chen
Kaixuan Shi

Movement initiation and termination rely on cortical circuits that must flexibly switch between active and quiescent states, a process markedly compromised in Parkinson's disease (PD). Although exercise has documented benefits in alleviating parkinsonian motor deficits, how it restores state-transition flexibility at the cortical population level remains insufficiently understood. Here, male C57BL/6 mice were assigned to control (CG), 6-hydroxydopamine-lesioned (PD), and PD with treadmill training (PDEX) groups. We combined free-moving behavioral tracking with in vivo single-unit and local field potential recordings from the right primary motor cortex (M1). We identified multiple transition-related firing patterns in M1, with distributions differing across groups. The proportion of biphasic neurons (start-excitation/stop-inhibition) was similar across CG (34.4%), PD (33.1%), and PDEX (36.9%) groups (p = 0.358). In contrast, the proportion of start-only neurons (start-excitation/stop-no response) was significantly increased in the PD group (23.5%) compared with CG (11.7%) and was intermediate in the PDEX group (17.9%) (p = 0.001). In parallel, CG exhibited transition-related beta power modulation during both initiation and termination, whereas this modulation was absent in PD animals. In the exercise group, transition-related modulation emerged in the gamma-band. Furthermore, waveform analysis showed that these response patterns were observed in both putative pyramidal neurons and interneurons. Together, these results show that exercise mitigates PD-related motor dysfunction, at least in part, by reinstating the diversity of M1 transition-related responses, highlighting cortical transition coding as a plastic and exercise-responsive target for functional recovery in PD.

Details DOI

NeurIPS Conference 2025 Conference Paper

Dimension-Reduction Attack! Video Generative Models are Experts on Controllable Image Synthesis

Hengyuan Cao
Yutong Feng
Biao Gong
Yijing Tian
Yunhong Lu
Chuang Liu
Bin Wang

Video generative models can be regarded as world simulators due to their ability to capture dynamic, continuous changes inherent in real-world environments. These models integrate high-dimensional information across visual, temporal, spatial, and causal dimensions, enabling predictions of subjects in various status. A natural and valuable research direction is to explore whether a fully trained video generative model in high-dimensional space can effectively support lower-dimensional tasks such as controllable image generation. In this work, we propose a paradigm for video-to-image knowledge compression and task adaptation, termed \textit{Dimension-Reduction Attack} (\texttt{DRA-Ctrl}), which utilizes the strengths of video models, including long-range context modeling and flatten full-attention, to perform various generation tasks. Specially, to address the challenging gap between continuous video frames and discrete image generation, we introduce a mixup-based transition strategy that ensures smooth adaptation. Moreover, we redesign the attention structure with a tailored masking mechanism to better align text prompts with image-level control. Experiments across diverse image generation tasks, such as subject-driven and spatially conditioned generation, show that repurposed video models outperform those trained directly on images. These results highlight the untapped potential of large-scale video generators for broader visual applications. \texttt{DRA-Ctrl} provides new insights into reusing resource-intensive video models and lays foundation for future unified generative models across visual modalities. The project page is \url{https: //dra-ctrl-2025. github. io/DRA-Ctrl/}.

PDF Details

AAAI Conference 2025 Conference Paper

Perturbating, Tuning, and Collaborating: Harnessing Vision Foundation Models for Single Domain Generalization on Medical Imaging

Chuang Liu
Yichao Cao
Yingying Zhang
Xiu Su
Haogang Zhu

Single Domain Generalization (SDG) is critical in medical imaging applications. Recently, Vision Foundation Models (VFMs) have spearheaded a trend in AI development due to their robust generalizability and versatility. This work aims to fully explore the generalization capabilities of VFMs alongside the domain-specific expertise of specialized models, thoroughly investigating the boundaries of their respective capabilities, thereby collaboratively addressing SDG challenges within medical imaging. We propose a framework for Collaborative reasoning between Specialized and Universal models for Single Domain Generalization (CollaSU-SDG) in medical imaging. Specifically, we first design a model-aware perturbation injection method from the perspective of single-source domain data, enabling differentiated and adaptive perturbation injection for two different scales of models. Then, a domain expansion adapter is designed for the VFM to adapt to the augmented single-source domain medical data. Lastly, we introduce an adaptive hierarchical transfer and dynamic dense prompting method that facilitate collaborative reasoning between the specialized and universal models, eliminating the need for explicit prompts. Through these designs, CollaSU-SDG fully leverages the strengths of both specialized and universal models, achieving robust out-of-distribution generalization capabilities on single-source domain data. Experimental results demonstrate that CollaSU-SDG significantly advances the state-of-the-art performance across a wide range of medical datasets. All the code will be publicly available.

PDF Details DOI

JBHI Journal 2025 Journal Article

The Devil is in the Frequency: Constrained and Adaptive Fine-Grained Domain Perturbation for Robust Medical Segmentation

Chuang Liu
Yichao Cao
Haogang Zhu

Domain generalization (DG) in medical image analysis is critical for achieving consistent and reliable diagnostics across diverse healthcare systems. However, domain shifts resulting from variations in imaging protocols, devices, and practices hinder accurate anatomical identification. While data augmentation shows promise, it struggles to generate diverse samples that bridge domain gaps and often distorts invariant anatomical features, compromising diagnostic integrity. This paper introduces the Adaptive Dual-Space Spectral Perturbation (AdaDSP) framework to address these issues at both broad and fine-grained levels. At the broad level, AdaDSP injects learnable spectral perturbations into input images and intermediate feature maps, significantly enhancing the diversity of the training data. At the fine-grained level, we propose a Fine-Grained Spectral Perturbation module that utilizes two lightweight attention mechanisms to capture sensitive frequency bands that hinder generalization. By injecting multivariate Gaussian noise within a mini-batch, this module better modulates the distribution of frequencies and accomplishes adaptive perturbation of sensitive frequency bands. Furthermore, we introduce a Universal Triple-stage Semantic Constraint Framework to encourage the networks to learn domain-invariant representations while retaining the discriminabtive capacity. Extensive experiments show that our method outperforms state-of-the-art benchmarks, with improvements of 2. 40% and 2. 99% in two notable medical imaging tasks, respectively.

Details DOI

ICML Conference 2025 Conference Paper

TinyMIG: Transferring Generalization from Vision Foundation Models to Single-Domain Medical Imaging

Chuang Liu
Hongyan Xu 0002
Yichao Cao
Xiu Su
Zhe Qu
Tianfa Li
Shan An
Haogang Zhu

Medical imaging faces significant challenges in single-domain generalization (SDG) due to the diversity of imaging devices and the variability among data collection centers. To address these challenges, we propose TinyMIG, a framework designed to transfer generalization capabilities from vision foundation models to medical imaging SDG. TinyMIG aims to enable lightweight specialized models to mimic the strong generalization capabilities of foundation models in terms of both global feature distribution and local fine-grained details during training. Specifically, for global feature distribution, we propose a Global Distribution Consistency Learning strategy that mimics the prior distributions of the foundation model layer by layer. For local fine-grained details, we further design a Localized Representation Alignment method, which promotes semantic alignment and generalization distillation between the specialized model and the foundation model. These mechanisms collectively enable the specialized model to achieve robust performance in diverse medical imaging scenarios. Extensive experiments on large-scale benchmarks demonstrate that TinyMIG, with extremely low computational cost, significantly outperforms state-of-the-art models, showcasing its superior SDG capabilities. All the code and model weights will be publicly available.

Details

IJCAI Conference 2024 Conference Paper

Gradformer: Graph Transformer with Exponential Decay

Chuang Liu
Zelin Yao
Yibing Zhan
Xueqi Ma
Shirui Pan
Wenbin Hu

Graph Transformers (GTs) have demonstrated their advantages across a wide range of tasks. However, the self-attention mechanism in GTs overlooks the graph's inductive biases, particularly biases related to structure, which are crucial for the graph tasks. Although some methods utilize positional encoding and attention bias to model inductive biases, their effectiveness is still suboptimal analytically. Therefore, this paper presents Gradformer, a method innovatively integrating GT with the intrinsic inductive bias by applying an exponential decay mask to the attention matrix. Specifically, the values in the decay mask matrix diminish exponentially, correlating with the decreasing node proximities within the graph structure. This design enables Gradformer to retain its ability to capture information from distant nodes while focusing on the graph's local details. Furthermore, Gradformer introduces a learnable constraint into the decay mask, allowing different attention heads to learn distinct decay masks. Such an design diversifies the attention heads, enabling a more effective assimilation of diverse structural information within the graph. Extensive experiments on various benchmarks demonstrate that Gradformer consistently outperforms the Graph Neural Network and GT baseline models in various graph classification and regression tasks. Additionally, Gradformer has proven to be an effective method for training deep GT models, maintaining or even enhancing accuracy compared to shallow models as the network deepens, in contrast to the significant accuracy drop observed in other GT models. Codes are available at https: //github. com/LiuChuang0059/Gradformer.

PDF Details DOI

IJCAI Conference 2024 Conference Paper

Where to Mask: Structure-Guided Masking for Graph Masked Autoencoders

Chuang Liu
Yuyao Wang
Yibing Zhan
Xueqi Ma
Dapeng Tao
Jia Wu
Wenbin Hu

Graph masked autoencoders (GMAE) have emerged as a significant advancement in self-supervised pre-training for graph-structured data. Previous GMAE models primarily utilize a straightforward random masking strategy for nodes or edges during training. However, this strategy fails to consider the varying significance of different nodes within the graph structure. In this paper, we investigate the potential of leveraging the graph's structural composition as a fundamental and unique prior in the masked pre-training process. To this end, we introduce a novel structure-guided masking strategy (i. e. , StructMAE), designed to refine the existing GMAE models. StructMAE involves two steps: 1) Structure-based Scoring: Each node is evaluated and assigned a score reflecting its structural significance. Two distinct types of scoring manners are proposed: predefined and learnable scoring. 2) Structure-guided Masking: With the obtained assessment scores, we develop an easy-to-hard masking strategy that gradually increases the structural awareness of the self-supervised reconstruction task. Specifically, the strategy begins with random masking and progresses to masking structure-informative nodes based on the assessment scores. This design gradually and effectively guides the model in learning graph structural information. Furthermore, extensive experiments consistently demonstrate that our StructMAE method outperforms existing state-of-the-art GMAE models in both unsupervised and transfer learning tasks. Codes are available at https: //github. com/LiuChuang0059/StructMAE.

PDF Details DOI

IJCAI Conference 2023 Conference Paper

Gapformer: Graph Transformer with Graph Pooling for Node Classification

Chuang Liu
Yibing Zhan
Xueqi Ma
Liang Ding
Dapeng Tao
Jia Wu
Wenbin Hu

Graph Transformers (GTs) have proved their advantage in graph-level tasks. However, existing GTs still perform unsatisfactorily on the node classification task due to 1) the overwhelming unrelated information obtained from a vast number of irrelevant distant nodes and 2) the quadratic complexity regarding the number of nodes via the fully connected attention mechanism. In this paper, we present Gapformer, a method for node classification that deeply incorporates Graph Transformer with Graph Pooling. More specifically, Gapformer coarsens the large-scale nodes of a graph into a smaller number of pooling nodes via local or global graph pooling methods, and then computes the attention solely with the pooling nodes rather than all other nodes. In such a manner, the negative influence of the overwhelming unrelated nodes is mitigated while maintaining the long-range information, and the quadratic complexity is reduced to linear complexity with respect to the fixed number of pooling nodes. Extensive experiments on 13 node classification datasets, including homophilic and heterophilic graph datasets, demonstrate the competitive performance of Gapformer over existing Graph Neural Networks and GTs.

PDF Details DOI

IJCAI Conference 2023 Conference Paper

Graph Pooling for Graph Neural Networks: Progress, Challenges, and Opportunities

Chuang Liu
Yibing Zhan
Jia Wu
Chang Li
Bo Du
Wenbin Hu
Tongliang Liu
Dacheng Tao

Graph neural networks have emerged as a leading architecture for many graph-level tasks, such as graph classification and graph generation. As an essential component of the architecture, graph pooling is indispensable for obtaining a holistic graph-level representation of the whole graph. Although a great variety of methods have been proposed in this promising and fast-developing research field, to the best of our knowledge, little effort has been made to systematically summarize these works. To set the stage for the development of future works, in this paper, we attempt to fill this gap by providing a broad review of recent methods for graph pooling. Specifically, 1) we first propose a taxonomy of existing graph pooling methods with a mathematical summary for each category; 2) then, we provide an overview of the libraries related to graph pooling, including the commonly used datasets, model architectures for downstream tasks, and open-source implementations; 3) next, we further outline the applications that incorporate the idea of graph pooling in a variety of domains; 4) finally, we discuss certain critical challenges facing current studies and share our insights on future potential directions for research on the improvement of graph pooling.

PDF Details DOI

NeurIPS Conference 2022 Conference Paper

TGEA 2.0: A Large-Scale Diagnostically Annotated Dataset with Benchmark Tasks for Text Generation of Pretrained Language Models

Huibin Ge
Xiaohu Zhao
Chuang Liu
Yulong Zeng
Qun Liu
Deyi Xiong

In order to diagnostically analyze and improve the capability of pretrained language models (PLMs) in text generation, we propose TGEA 2. 0, to date the largest dataset built on machine-authored texts by PLMs with fine-grained semantic annotations on a wide variety of pathological generation errors. We collect 170K nominal, phrasal and sentential prompts from 6M natural sentences in 3 domains. These prompts are fed into 4 generative PLMs with their best decoding strategy to generate paragraphs. 195, 629 sentences are extracted from these generated paragraphs for manual annotation, where 36K erroneous sentences are detected, 42K erroneous spans are located and categorized into an error type defined in a two-level error taxonomy. We define a \textbf{Mi}nimal \textbf{S}et of \textbf{E}rror-related \textbf{W}ords (MiSEW) for each erroneous span, which not only provides error-associated words but also rationalizes the reasoning behind the error. Quality control with a pre-annotation and feedback loop is performed before and during the entire annotation process. With the diagnostically annotated dataset, we propose 5 diagnosis benchmark tasks (i. e. , erroneous text detection, MiSEW extraction, erroneous span location and correction together with error type classification) and 2 pathology mitigation benchmark tasks (pairwise comparison and word prediction). Experiment results on these benchmark tasks demonstrate that TGEA 2. 0 is a challenging dataset that could facilitate further research on automatic diagnosis and pathology mitigation over machine texts. The dataset will be publicly available at https: //github. com/tjunlp-lab/TGEA/.

PDF Details

TCS Journal 2020 Journal Article

Group sweep coverage with guaranteed approximation ratio

Chuang Liu
Hongwei Du
Qiang Ye
Wen Xu

Wireless Sensor Networks (WSNs) are often deployed to monitor a region of interest. With sweep coverage, mobile sensor nodes are scheduled to move along a planned route (i. e. sweep route) in order to collect the data from a series of Point of Interests (POIs) sequentially. In this paper, we generalize the sweep coverage problem by proposing a new coverage paradigm, group sweep coverage. With group sweep coverage, the POIs are divided into several groups. A group is said to be covered when one of the POIs in the group is covered. The goal in group sweep coverage is to construct a sweep route that mobile sensor nodes should follow in order to cover all groups during each predefined period. In our research, we devised two algorithms for group sweep coverage: AGSC and DSRM. AGSC is a centralized scheme whose approximation ratio is 5Δ. Namely, the length of the sweep route generated by AGSC is at most 5Δ times that of the optimal sweep route. DSRM is a distributed scheme for large-scale networks with dynamic POIs. Compared with AGSC, DSRM leads to the same approximation ratio and better scalability. Our experimental results indicate that both AGSC and DSRM outperform the state-of-the-art schemes in terms of average and maximal sweep route length.

Details DOI

YNIMG Journal 2004 Journal Article

Attenuation of brain response to heroin correlates with the reinstatement of heroin-seeking in rats by fMRI

Feng Luo
Zheng-Xiong Xi
Gaohong Wu
Chuang Liu
Eliot L Gardner
Shi-Jiang Li

Thirty male Sprague–Dawley rats were divided into two groups and trained to self-administer either saline (n = 14) or heroin (0. 1 mg/kg per injection, n = 16) for 10–12 days until a stable self-administration (SA) behavior was achieved. After 8–9 days of withdrawal, each group was divided into two subgroups for reinstatement tests and functional magnetic resonance image (fMRI) scanning, respectively, to determine the neural correlates of the reinstatement of heroin-seeking behavior. For reinstatement testing, heroin-SA rats (n = 10) displayed robust reinstatement of drug-seeking behavior triggered by an acute heroin priming injection, whereas saline control rats (n = 8) did not show such a behavioral response. Regional positive or negative blood oxygen level-dependent (BOLD) signals, induced by heroin priming injection, were observed in both groups of rats during fMRI scanning. However, such heroin-induced positive BOLD signal primarily in the prefrontal cortex and parietal cortex was significantly attenuated in heroin-SA rats (n = 6) when compared to saline control rats (n = 6). Similarly, the heroin-induced negative BOLD signal in the subcortical regions, such as in the nucleus accumbens and hippocampus, was also significantly attenuated in both signal intensity and number of brain voxels activated in heroin-SA rats. These data demonstrate that heroin-induced reinstatement of drug-seeking behavior coincides with a significant, enduring reduction in opiate-induced brain activity in heroin-SA rats, suggesting a possible role of opiate tolerance in mediating reinstatement of drug-seeking behavior.

Details DOI