Wenyu Chen

AAAI Conference 2026 Conference Paper

HardF-SNN: Hardware-Friendly Quantization for Spiking Neural Networks with Efficient Integer-Arithmetic-Only Inference

• Hanwen Liu
• Kexin Shi
• Jieyuan Zhang
• Yimeng Shan
• Jibin Wu
• Wenyu Chen
• Malu Zhang

Spiking Neural Networks (SNNs) are emerging as a promising energy-efficient alternative to Artificial Neural Networks (ANNs) due to their event-driven computation paradigm. However, recent advances toward large-scale high-performance SNNs inevitably lead to substantial memory and computational overhead. While quantization offers a potential way, many quantization approaches fail to deliver verifiable efficiency gains on resource-constrained hardware platforms. In this paper, we propose a lightweight and hardware-friendly SNN, termed HardF-SNN. Specifically, we first build a baseline model using shared-scale quantization and BN folding to simulate integer-only inference, as this has not been thoroughly discussed in prior SNN works. Then, through empirical and theoretical analysis, we identify that the baseline suffers from accuracy degradation and may cause training failure. To mitigate these issues, we propose proportional shared-scale quantization for enhanced dynamic range and integer-only BN using bit-shifting to stabilize training. Extensive experiments show that HardF-SNN achieves an optimal balance between performance and efficiency with excellent hardware compatibility. To demonstrate its effectiveness on resource-limited platforms, HardF-SNN is deployed on a dedicated FPGA-based hardware accelerator. Evaluation results indicate that our implementation achieves significant performance improvements over several existing hardware accelerators.

NeurIPS Conference 2025 Conference Paper

Beyond Pairwise Connections: Extracting High-Order Functional Brain Network Structures under Global Constraints

• Ling Zhan
• Junjie Huang
• Xiaoyao Yu
• Wenyu Chen
• Tao Jia

Functional brain network (FBN) modeling often relies on local pairwise interactions, whose limitation in capturing high-order dependencies is theoretically analyzed in this paper. Meanwhile, the computational burden and heuristic nature of current hypergraph modeling approaches hinder end-to-end learning of FBN structures directly from data distributions. To address this, we propose to extract high-order FBN structures under global constraints, and implement this as a Global Constraints oriented Multi-resolution (GCM) FBN structure learning framework. It incorporates 4 types of global constraint (signal synchronization, subject identity, expected edge numbers, and data labels) to enable learning FBN structures for 4 distinct levels (sample/subject/group/project) of modeling resolution. Experimental results demonstrate that GCM achieves up to a 30. 6% improvement in relative accuracy and a 96. 3% reduction in computational time across 5 datasets and 2 task settings, compared to 9 baselines and 10 state-of-the-art methods. Extensive experiments validate the contributions of individual components and highlight the interpretability of GCM. This work offers a novel perspective on FBN structure learning and provides a foundation for interdisciplinary applications in cognitive neuroscience. Code is publicly available on https: //github. com/lzhan94swu/GCM.

ICML Conference 2025 Conference Paper

Diffusion Sampling Correction via Approximately 10 Parameters

• Guangyi Wang
• Wei Peng 0009
• Lijiang Li
• Wenyu Chen
• Yuren Cai
• Songzhi Su

While powerful for generation, Diffusion Probabilistic Models (DPMs) face slow sampling challenges, for which various distillation-based methods have been proposed. However, they typically require significant additional training costs and model parameter storage, limiting their practicality. In this work, we propose P CA-based A daptive S earch (PAS), which optimizes existing solvers for DPMs with minimal additional costs. Specifically, we first employ PCA to obtain a few basis vectors to span the high-dimensional sampling space, which enables us to learn just a set of coordinates to correct the sampling direction; furthermore, based on the observation that the cumulative truncation error exhibits an “S"-shape, we design an adaptive search strategy that further enhances the sampling efficiency and reduces the number of stored parameters to approximately 10. Extensive experiments demonstrate that PAS can significantly enhance existing fast solvers in a plug-and-play manner with negligible costs. E. g. , on CIFAR10, PAS optimizes DDIM’s FID from 15. 69 to 4. 37 (NFE=10) using only 12 parameters and sub-minute training on a single A100 GPU. Code is available at https: //github. com/onefly123/PAS.

NeurIPS Conference 2025 Conference Paper

ErrorTrace: A Black-Box Traceability Mechanism Based on Model Family Error Space

• Chuanchao Zang
• Xiangtao Meng
• Wenyu Chen
• Tianshuo Cong
• Zha Yaxing
• Dong Qi
• Zheng Li
• Shanqing Guo

The open-source release of large language models (LLMs) enables malicious users to create unauthorized derivative models at low cost, posing significant threats to intellectual property (IP) and market stability. Existing IP protection methods either require access to model parameters or are vulnerable to fine-tuning attacks. To fill this gap, we propose ErrorTrace, a robust and black-box traceability mechanism for protecting LLM IP. Specifically, ErrorTrace leverages the unique error patterns of model families by mapping and analyzing their distinct error spaces, enabling robust and efficient IP protection without relying on internal parameters or specific query responses. Experimental results show that ErrorTrace achieves a traceability accuracy of 0. 8518 for 27 base models when the suspect model is not included in ErrorTrace's training set, outperforming the baseline by 0. 2593. Additionally, ErrorTrace successfully tracks 34 fine-tuned, pruned and merged models across various scenarios, demonstrating its broad applicability and robustness. In addition, ErrorTrace shows a certain level of resilience when subjected to adversarial attacks. Our code is available at: https: //github. com/csdatazcc/ErrorTrace.

AAAI Conference 2025 Conference Paper

Leveraging Asynchronous Spiking Neural Networks for Ultra Efficient Event-Based Visual Processing

• DingYi Zeng
• Yuchen Wang
• Honglin Cao
• Wanlong Liu
• Yichen Xiao
• ChengzhuoLu
• Wenyu Chen
• Malu Zhang

Event cameras encode visual information by generating asynchronous and sparse event streams, which hold great potential for low latency and low power consumption. Despite many successful implementations of event camera-based applications, most of them accumulate the events into frames and then utilize conventional frame-based computer vision algorithms. These frame-based methods, though typically effective, diminish the inherent advantages of the event camera's low latency and low power consumption. To solve the above problems, we propose ASGCN, which efficiently processes data on an event-by-event basis and dynamically evolves into a corresponding dynamic representation, enabling low latency and high sparsity of data representation. The sparsity computation is further improved by introducing brain-inspired spiking neural networks, resulting in low power consumption for ASGCN. Extensive and diverse experiments demonstrate the energy efficiency and low latency advantages of our processing pipeline. Especially on real-world event camera datasets, our pipeline consumes more than 10,000 times less energy and achieves similar performance compared to current frame-based methods.

AAAI Conference 2025 Conference Paper

One Node One Model: Featuring the Missing-Half for Graph Clustering

• Xuanting Xie
• Bingheng Li
• Erlin Pan
• Zhaochen Guo
• Zhao Kang
• Wenyu Chen

Most existing graph clustering methods primarily focus on exploiting topological structure, often neglecting the "missing-half" node feature information, especially how these features can enhance clustering performance. This issue is further compounded by the challenges associated with high-dimensional features. Feature selection in graph clustering is particularly difficult because it requires simultaneously discovering clusters and identifying the relevant features for these clusters. To address this gap, we introduce a novel paradigm called "one node one model", which builds an exclusive model for each node and defines the node label as a combination of predictions for node groups. Specifically, the proposed "Feature Personalized Graph Clustering (FPGC)" method identifies cluster-relevant features for each node using a squeeze-and-excitation block, integrating these features into each model to form the final representations. Additionally, the concept of feature cross is developed as a data augmentation technique to learn low-order feature interactions. Extensive experimental results demonstrate that FPGC outperforms state-of-the-art clustering methods. Moreover, the plug-and-play nature of our method provides a versatile solution to enhance GNN-based models from the feature perspective.

NeurIPS Conference 2025 Conference Paper

QFFT, Question-Free Fine-Tuning for Adaptive Reasoning

• Wanlong Liu
• Junxiao Xu
• Fei Yu
• Yukang Lin
• Ke Ji
• Wenyu Chen
• Lifeng Shang
• Yasheng Wang

Recent advancements in Long Chain-of-Thought (CoT) reasoning models have improved performance on complex tasks, but they suffer from overthinking, which generates redundant reasoning steps, especially for simple questions. This paper revisits the reasoning patterns of Long and Short CoT models, observing that the Short CoT patterns offer concise reasoning efficiently, while the Long CoT patterns excel in challenging scenarios where the Short CoT patterns struggle. To enable models to leverage both patterns, we propose Question-Free Fine-Tuning (QFFT), a fine-tuning approach that removes the input question during training and learns exclusively from Long CoT responses. This approach enables the model to adaptively employ both reasoning patterns: it prioritizes the Short CoT patterns and activates the Long CoT patterns only when necessary. Experiments on various mathematical datasets demonstrate that QFFT reduces average response length by more than 50\%, while achieving performance comparable to Supervised Fine-Tuning (SFT). Additionally, QFFT exhibits superior performance compared to SFT in noisy, out-of-domain, and low-resource scenarios.

YNIMG Journal 2025 Journal Article

Retraction notice to “Multi-feature concatenation and multi-classifier stacking: An interpretable and generalizable machine learning method for MDD discrimination with rsfMRI” [NeuroImage 285 (2024) 120497]

• Yunsong Luo
• Wenyu Chen
• Ling Zhan
• Jiang Qiu
• Tao Jia

IROS Conference 2024 Conference Paper

3D Object Visibility Prediction in Autonomous Driving

• Chuanyu Luo
• Nuo Cheng
• Ren Zhong
• Haipeng Jiang
• Wenyu Chen
• Aoli Wang
• Pu Li 0001

With the rapid advancement of hardware and software technologies, research in autonomous driving has seen significant growth. The prevailing framework for multi-sensor autonomous driving encompasses sensor installation, perception, path planning, decision-making, and motion control. At the perception phase, a common approach involves utilizing neural networks to infer 3D bounding box (Bbox) attributes from raw sensor data, including classification, size, and orientation. In this paper, we present a novel attribute and its corresponding algorithm: 3D object visibility. By incorporating multi-task learning, the introduction of this attribute, visibility, negligibly affects the model's effectiveness and efficiency. Our proposal of this attribute and its computational strategy aims to expand the capabilities for downstream tasks, thereby enhancing the safety and reliability of real-time autonomous driving in real-world scenarios.

YNIMG Journal 2024 Journal Article

RETRACTED: Multi-feature concatenation and multi-classifier stacking: An interpretable and generalizable machine learning method for MDD discrimination with rsfMRI

• Yunsong Luo
• Wenyu Chen
• Ling Zhan
• Jiang Qiu
• Tao Jia

Major depressive disorder (MDD) is a serious and heterogeneous psychiatric disorder that needs accurate diagnosis. Resting-state functional MRI (rsfMRI), which captures multiple perspectives on brain structure, function, and connectivity, is increasingly applied in the diagnosis and pathological research of MDD. Different machine learning algorithms are then developed to exploit the rich information in rsfMRI and discriminate MDD patients from normal controls. Despite recent advances reported, the MDD discrimination accuracy has room for further improvement. The generalizability and interpretability of the discrimination method are not sufficiently addressed either. Here, we propose a machine learning method (MFMC) for MDD discrimination by concatenating multiple features and stacking multiple classifiers. MFMC is tested on the REST-meta-MDD data set that contains 2428 subjects collected from 25 different sites. MFMC yields 96.9% MDD discrimination accuracy, demonstrating a significant improvement over existing methods. In addition, the generalizability of MFMC is validated by the good performance when the training and testing subjects are from independent sites. The use of XGBoost as the meta classifier allows us to probe the decision process of MFMC. We identify 13 feature values related to 9 brain regions including the posterior cingulate gyrus, superior frontal gyrus orbital part, and angular gyrus, which contribute most to the classification and also demonstrate significant differences at the group level. The use of these 13 feature values alone can reach 87% of MFMC's full performance when taking all feature values. These features may serve as clinically useful diagnostic and prognostic biomarkers for MDD in the future.

AAAI Conference 2023 Conference Paper

Substructure Aware Graph Neural Networks

• DingYi Zeng
• Wanlong Liu
• Wenyu Chen
• Li Zhou
• Malu Zhang
• Hong Qu

Despite the great achievements of Graph Neural Networks (GNNs) in graph learning, conventional GNNs struggle to break through the upper limit of the expressiveness of first-order Weisfeiler-Leman graph isomorphism test algorithm (1-WL) due to the consistency of the propagation paradigm of GNNs with the 1-WL.Based on the fact that it is easier to distinguish the original graph through subgraphs, we propose a novel framework neural network framework called Substructure Aware Graph Neural Networks (SAGNN) to address these issues. We first propose a Cut subgraph which can be obtained from the original graph by continuously and selectively removing edges. Then we extend the random walk encoding paradigm to the return probability of the rooted node on the subgraph to capture the structural information and use it as a node feature to improve the expressiveness of GNNs. We theoretically prove that our framework is more powerful than 1-WL, and is superior in structure perception. Our extensive experiments demonstrate the effectiveness of our framework, achieving state-of-the-art performance on a variety of well-proven graph tasks, and GNNs equipped with our framework perform flawlessly even in 3-WL failed graphs. Specifically, our framework achieves a maximum performance improvement of 83% compared to the base models and 32% compared to the previous state-of-the-art methods.

AAAI Conference 2021 Conference Paper

Self-Paced Two-dimensional PCA

• Jiangxin Li
• Zhao Kang
• Chong Peng
• Wenyu Chen

Two-dimensional PCA (2DPCA) is an effective approach to reduce dimension and extract features in the image domain. Most recently developed techniques use different error measures to improve their robustness to outliers. When certain data points are overly contaminated, the existing methods are frequently incapable of filtering out and eliminating the excessively polluted ones. Moreover, natural systems have smooth dynamics, an opportunity is lost if an unsupervised objective function remains static. Unlike previous studies, we explicitly differentiate the samples to alleviate the impact of outliers and propose a novel method called Self-Paced 2DPCA (SP2DPCA) algorithm, which progresses from ‘easy’ to ‘complex’ samples. By using an alternative optimization strategy, SP2DPCA looks for optimal projection matrix and filters out outliers iteratively. Theoretical analysis demonstrates the robustness nature of our method. Extensive experiments on image reconstruction and clustering verify the superiority of our approach.

IJCAI Conference 2019 Conference Paper

Multiple Partitions Aligned Clustering

• Zhao Kang
• Zipeng Guo
• Shudong Huang
• Siying Wang
• Wenyu Chen
• Yuanzhang Su
• Zenglin Xu

Multi-view clustering is an important yet challenging task due to the difficulty of integrating the information from multiple representations. Most existing multi-view clustering methods explore the heterogeneous information in the space where the data points lie. Such common practice may cause significant information loss because of unavoidable noise or inconsistency among views. Since different views admit the same cluster structure, the natural space should be all partitions. Orthogonal to existing techniques, in this paper, we propose to leverage the multi-view information by fusing partitions. Specifically, we align each partition to form a consensus cluster indicator matrix through a distinct rotation matrix. Moreover, a weight is assigned for each view to account for the clustering capacity differences of views. Finally, the basic partitions, weights, and consensus clustering are jointly learned in a unified framework. We demonstrate the effectiveness of our approach on several real datasets, where significant improvement is found over other state-of-the-art multi-view clustering methods.

JBHI Journal 2016 Journal Article

Volume Preserved Mass–Spring Model with Novel Constraints for Soft Tissue Deformation

• Yuping Duan
• Weimin Huang
• Huibin Chang
• Wenyu Chen
• Jiayin Zhou
• Soo Kng Teo
• Yi Su
• Chee Kong Chui

An interactive surgical simulation system needs to meet three main requirements, speed, accuracy, and stability. In this paper, we present a stable and accurate method for animating mass–spring systems in real time. An integration scheme derived from explicit integration is used to obtain interactive realistic animation for a multiobject environment. We explore a predictor–corrector approach by correcting the estimation of the explicit integration in a poststep process. We introduce novel constraints on positions into the mass–spring model (MSM) to model the nonlinearity and preserve volume for the realistic simulation of the incompressibility. We verify the proposed MSM by comparing its deformations with the reference deformations of the nonlinear finite-element method. Moreover, experiments on porcine organs are designed for the evaluation of the multiobject deformation. Using a pair of freshly harvested porcine liver and gallbladder, the real organ deformations are acquired by computed tomography and used as the reference ground truth. Compared to the porcine model, our model achieves a $1. 502$ mm mean absolute error measured at landmark locations for cases with small deformation (the largest deformation is $49. 109$ mm) and a $3. 639$ mm mean absolute error for cases with large deformation (the largest deformation is $83. 137$ mm). The changes of volume for the two deformations are limited to $0. 030\%$ and $0. 057\%$, respectively. Finally, an implementation in a virtual reality environment for laparoscopic cholecystectomy demonstrates that our model is capable to simulate large deformation and preserve volume in real-time calculations.