Author name cluster

Lin Zhang

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.

33 papers

2 author rows

EAAI Journal 2026 Journal Article

BoA-SQL: Executable Blueprint-of-Action for Text-to-SQL with reinforcement learning

Yang Wang
Zhilong Xie
Lin Zhang
Lingyun Gu
Qing Li

Recent progress in large language models has opened new possibilities for querying databases using natural language instead of SQL. Yet existing methods, often relying on linear reasoning, struggle with complex nested logic and are susceptible to error propagation. We propose BoA-SQL, which turns reasoning into a durable, structured plan. Our contributions are threefold: (1) To resolve schema ambiguity and context overload, a lightweight, knowledge graph-driven linker grounds the query by pruning irrelevant schema before planning. (2) To overcome the structural mismatch of linear plans, a persistent, tree-structured blueprint aligns with SQL’s hierarchy, enabling localized repair of faulty segments without full re-computation. (3) To align the language model’s text-generation objective with execution correctness, a two-stage reinforcement learning policy optimizes the entire blueprint for task success. Extensive evaluations on public benchmarks Spider and a complex real-world database validate this approach. BoA-SQL achieves 85. 6% execution accuracy. This figure notably increases to 88. 2% after correcting benchmark label errors. These findings suggest that durable, structure-aware planning, combined with schema grounding and planning-aligned optimization, is a practical path to reliable Text-to-SQL.

Details DOI

JBHI Journal 2026 Journal Article

GAEA-Net: Generating Activity-Enriched Abnormal ECGs via Adversarial Network

Liuqing Chen
Shuhong Xiao
Yujie Zang
Jiner Wang
Shanhai Hu
Lin Zhang
Qing Li
Danyang Zhang

With the increasing demand for personalized health monitoring through wearable devices, there is a growing need for non-prescription ECG diagnosing, especially during physical activities. However, existing abnormal ECG data are typically measured in clinical settings, reflecting heart waveforms in a resting state. Abnormality classification models based on such data often struggle to maintain high performance during physical activities, leading to increased false alarms and a higher incidence of missed detections. Due to the potential risks associated with having patients engage in physical activity, abnormal ECG data captured during exercise is not readily available, further complicating the development of reliable models for active scenarios. To address this issue, we propose GAEA-Net in this study. Our goal is to utilize exercise ECGs from healthy individuals, which are more easily accessible, combined with resting-state abnormal ECGs, to generate activity-enriched ECGs through synthesis. We conduct abnormal classification on five widely used datasets, achieving average improvements of 1. 3% in Accuracy, 1. 3% in F1-score, 0. 9% in AUROC, 1. 6% in MCC, and 1. 4% in Cohen’s Kappa. Furthermore, a clinical Turing test involving seven experienced cardiologists confirms that our synthesized ECGs exhibit high fidelity. In the diagnostic task, the cardiologists achieved comparable accuracy on synthetic and real ECGs (55. 7% vs. 54. 9%, p = 0. 76).

Details DOI

AAAI Conference 2026 Conference Paper

RealRep: Generalized SDR-to-HDR Conversion via Attribute-Disentangled Representation Learning

Li Xu
Siqi Wang
Kepeng Xu
Lin Zhang
Gang He
Weiran Wang
Yu-Wing Tai

High-Dynamic-Range Wide-Color-Gamut (HDR-WCG) technology is becoming increasingly widespread, driving a growing need for converting Standard Dynamic Range (SDR) content to HDR. Existing methods primarily rely on fixed tone mapping operators, which struggle to handle the diverse appearances and degradations commonly present in real-world SDR content. To address this limitation, we propose a generalized SDR-to-HDR framework that enhances robustness by learning attribute-disentangled representations. Central to our approach is Realistic Attribute-Disentangled Representation Learning (RealRep), which explicitly disentangles luminance and chrominance components to capture intrinsic content variations across different SDR distributions. Furthermore, we design a Luma-/Chroma-aware negative exemplar generation strategy that constructs degradation-sensitive contrastive pairs, effectively modeling tone discrepancies across SDR styles. Building on these attribute-level priors, we introduce the Degradation-Domain Aware Controlled Mapping Network (DDACMNet), a lightweight, two-stage framework that performs adaptive hierarchical mapping guided by a control-aware normalization mechanism. DDACMNet dynamically modulates the mapping process via degradation-conditioned features, enabling robust adaptation across diverse degradation domains. Extensive experiments demonstrate that RealRep consistently outperforms state-of-the-art methods in both generalization and perceptually faithful HDR color gamut reconstruction.

PDF Details DOI

AAAI Conference 2026 Conference Paper

SEBSFormer: A Spectral-Enhanced Bi-Stream Transformer for Robust EEG Decoding

Lin Zhang
Shikui Tu
Lei Xu

Electroencephalography (EEG) plays a vital role in clinical and cognitive applications such as epilepsy diagnosis and emotion recognition. However, the low signal-to-noise ratio, inter-subject variability, and inherent non-stationarity of EEG signals present substantial modeling challenges. While recent Transformer-based models offer promising long-range modeling capabilities, their self-attention mechanism behaves as a low-pass filter, suppressing high-frequency neural patterns critical for decoding transient events. In this work, we provide the first formal analysis demonstrating this low-pass behavior in self-attention mechanisms when applied to EEG signals, revealing a fundamental limitation of deep attention-based EEG models. To address this, we propose SEBSFormer, a spectral-enhanced bi-Stream Transformer that jointly models temporal dependencies and spectral structures. SEBSFormer integrates three key modules: a spectral compensation module that restores high-frequency components via residual correction in the Fourier domain; a multi-scale temporal attention module for saliency-guided temporal compression; and a graph-guided dynamic fusion module for adaptive spatial aggregation across electrodes. Extensive experiments on three benchmark datasets—TUAB, TUEV, and SEED—demonstrate that SEBSFormer consistently outperforms existing state-of-the-art models across both clinical and affective tasks. Our findings establish a new paradigm for frequency-aware EEG modeling.

PDF Details DOI

AAAI Conference 2026 Conference Paper

SmartSplat: Feature-Smart Gaussians for Scalable Compression of Ultra-High-Resolution Images

Linfei Li
Lin Zhang
Zhong Wang
Ying Shen

Recent advances in generative AI have accelerated the production of ultra-high-resolution visual content. However, traditional image formats face significant limitations in efficient compression and real-time decoding, which restricts their applicability on end-user devices. Inspired by 3D Gaussian Splatting, 2D Gaussian image models have achieved notable progress in enhancing image representation efficiency and quality. Nevertheless, existing methods struggle to balance compression ratios and reconstruction fidelity in ultra-high-resolution scenarios. To address these challenges, we propose SmartSplat, a highly adaptive and feature-aware GS-based image compression framework that effectively supports arbitrary image resolutions and compression ratios. By leveraging image-aware features such as gradients and color variances, SmartSplat introduces a Gradient-Color Guided Variational Sampling strategy alongside an Exclusion-based Uniform Sampling scheme, significantly improving the non-overlapping coverage of Gaussian primitives in pixel space. Additionally, a Scale-Adaptive Gaussian Color Sampling method is proposed to enhance the initialization of Gaussian color attributes across scales. Through joint optimization of spatial layout, scale, and color initialization, SmartSplat can efficiently capture both local structures and global textures of images using a limited number of Gaussians, achieving superior reconstruction quality under high compression ratios. Extensive experiments on DIV8K and a newly created 16K dataset demonstrate that SmartSplat significantly outperforms state-of-the-art methods at comparable compression ratios and surpasses their compression limits, exhibiting strong scalability and practical applicability. This framework can effectively alleviate the storage and transmission burdens of ultra-high-resolution images, providing a robust foundation for future high-efficiency visual content processing.

PDF Details DOI

EAAI Journal 2025 Journal Article

A physics-informed neural network for nonlinear deflection prediction of Ionic Polymer-Metal Composite based on Kolmogorov-Arnold networks

Lin Zhang
Lei Chen
Fuxiang An
Zixuan Peng
Yuhang Yang
Tingting Peng
Yongshi Song
Yanzheng Zhao

To characterize the highly nonlinear state response of the Ionic Polymer-Metal Composite (IPMC), coupled with inherent response uncertainties, we reformulate differential equations that encapsulate the highly nonlinear deflection of IPMC, accounting for uncertainties. Subsequently, we propose a data-driven approach utilizing a physics-informed neural network (PINN) to effectively solve this differential equation and predict the nonlinear deflection of IPMC actuators. To enhance the capacity for accurately capturing the intricate nonlinear behavior of IPMC, this study introduces an improved data-driven methodology that integrates Kolmogorov-Arnold Networks into the framework of PINNs (KAN-PINN). Through a series of experimental analyses and simulation studies, we demonstrate that KAN-PINN significantly outperforms traditional multilayer perceptron-based PINN (MLP-PINN), achieving a prediction accuracy improvement of 51. 85% for Ag-IPMC and 27. 54% for Pt-IPMC, along with a reduction in the convergence cycle. This research underscores the advantages of employing a data-driven, PINN-based approach in addressing the complexities associated with the nonlinear deflection of IPMC actuators. Moreover, it provides a substantial reference for future investigations and applications of IPMC.

Details DOI

AAAI Conference 2025 Conference Paper

BigMac: A Communication-Efficient Mixture-of-Experts Model Structure for Fast Training and Inference

Zewen Jin
Shengnan Wang
Jiaan Zhu
Hongrui Zhan
Youhui Bai
Lin Zhang
Zhenyu Ming
Cheng Li

The Mixture-of-Experts (MoE) structure scales the Transformer-based large language models (LLMs) and improves their performance with only the sub-linear increase in computation resources. Recently, a fine-grained DeepSeekMoE structure is proposed, which can further improve the computing efficiency of MoE without performance degradation. However, the All-to-All communication introduced by MoE has become a bottleneck, especially for the fine-grained structure, which typically involves and activates more experts, hence contributing to heavier communication overhead. In this paper, we propose a novel MoE structure named BigMac, which is also fine-grained but with high communication efficiency. The innovation of BigMac is mainly due to that we abandon the Communicate-Descend-Ascend-Communicate (CDAC) manner used by fine-grained MoE, which leads to the All-to-All communication always taking place at the highest dimension. Instead, BigMac designs an efficient Descend-Communicate-Communicate-Ascend (DCCA) manner. Specifically, we add a descending and ascending projection at the entrance and exit of the expert, respectively, which enables the communication to perform at a very low dimension. Furthermore, to adapt to DCCA, we re-design the structure of small experts, ensuring that the expert in BigMac has enough complexity to address tokens. Experimental results show that BigMac achieves comparable or even better model quality than fine-grained MoEs with the same number of experts and a similar number of total parameters. Equally importantly, BigMac reduces the end-to-end latency by up to 3.09 x for training and increases the throughput by up to 3.11 x for inference on state-of-the-art AI computing frameworks including Megatron, Tutel, and DeepSpeed-Inference.

PDF Details DOI

NeurIPS Conference 2025 Conference Paper

EAP-GP: Mitigating Saturation Effect in Gradient-based Automated Circuit Identification

Lin Zhang
Wenshuo Dong
Zhuoran Zhang
Shu Yang
Lijie Hu
Ninghao Liu
Pan Zhou
Di Wang

Understanding the internal mechanisms of transformer-based language models remains challenging. Mechanistic interpretability based on circuit discovery aims to reverse engineer neural networks by analyzing their internal processes at the level of computational subgraphs. In this paper, we revisit existing gradient-based circuit identification methods and find that their performance is either affected by the zero-gradient problem or saturation effects, where edge attribution scores become insensitive to input changes, resulting in noisy and unreliable attribution evaluations for circuit components. To address the saturation effect, we propose Edge Attribution Patching with GradPath (EAP-GP), EAP-GP introduces an integration path, starting from the input and adaptively following the direction of the difference between the gradients of corrupted and clean inputs to avoid the saturated region. This approach enhances attribution reliability and improves the faithfulness of circuit identification. We evaluate EAP-GP on 6 datasets using GPT-2 Small, GPT-2 Medium, and GPT-2 XL. Experimental results demonstrate that EAP-GP outperforms existing methods in circuit faithfulness, achieving improvements up to 17. 7\%. Comparisons with manually annotated ground-truth circuits demonstrate that EAP-GP achieves precision and recall comparable to or better than previous approaches, highlighting its effectiveness in identifying accurate circuits.

PDF Details

NeurIPS Conference 2025 Conference Paper

FAVOR-Bench: A Comprehensive Benchmark for Fine-Grained Video Motion Understanding

Chongjun Tu
Lin Zhang
pengtao chen
Peng Ye
Xianfang Zeng
Wei Cheng
Gang Yu
Tao Chen

Multimodal Large Language Models (MLLMs) have shown impressive video content understanding capabilities but struggle with fine-grained motion comprehension. To comprehensively assess the motion understanding ability of existing MLLMs, we introduce FAVOR-Bench, which comprises 1, 776 videos from both ego-centric and third-person perspectives and enables assessment through both close-ended and open-ended tasks. For close-ended evaluation, we carefully design 8, 184 multiple-choice question-answer pairs spanning six distinct sub-tasks. For open-ended evaluation, we employ the GPT-assisted evaluation and develop a novel cost-efficient LLM-free assessment method, where the latter can enhance benchmarking interpretability and accessibility. Comprehensive experiments with21 state-of-the-art MLLMs reveal significant limitations in their ability to comprehend and describe detailed temporal dynamics in video motions. To alleviate this limitation, we further build FAVOR-Train, a dataset of 17, 152 videos with fine-grained motion annotations. Finetuning Qwen2. 5-VL on FAVOR-Train yields consistent improvements on motion-related tasks across TVBench, MotionBenchand our FAVOR-Bench. Our assessment results demonstrate that the proposed FAVOR-Bench and FAVOR-Train provide valuable tools for the community to develop more powerful video understanding models.

PDF Details

JBHI Journal 2025 Journal Article

FDDSeg: Unleashing the Power of Scribble Annotation for Cardiac MRI Images Through Feature Decomposition Distillation

Lin Zhang
Wenzong Li
Kaiyue Bi
Pei Li
Liang Zhang
Hui Liu

Cardiovascular diseases can be diagnosed with computer assistance when using the magnetic resonance imaging (MRI) image that is produced by the MRI sensor. Deep learning-based scribbling MRI image segmentation has demonstrated impressive results recently. However, the majority of current approaches possess an excessive number of model parameters and do not completely utilize scribbling annotations. To develop a feature decomposition distillation deep learning method, named FDDSeg, for scribble-supervised cardiac MRI image segmentation. Public ACDC and MSCMR cardiac MRI datasets were used to evaluate the segmentation performance of FDDSeg. FDDSeg adopts a scribble annotation reuse policy to help provide accurate boundaries, and the intermediate features are split class region and class-free region by using the pseudo labels to further improve feature learning. Effective distillation knowledge is then captured by feature decomposition. FDDSeg was compared with 7 state-of-the-art methods, MAAG, ShapePU, CycleMix, Dual-Branch, ZscribbleSeg, Perturbation Dual-Branch as well as ScribbleVC on both ACDC and MSCMR datasets. FDDSeg is shown to perform the best in DSC(89. 05% and 88. 75%), JC(80. 30% and 79. 78%) as well as HD95(5. 76% and 4. 44%) metrics with only 2. 01 M of parameters. FDDSeg methods can segment cardiac MRI images more precise with only scribble annotations at lower computation cost, which may help increase the efficiency of quantitative analysis of cardiac.

Details DOI

NeurIPS Conference 2025 Conference Paper

PaceLLM: Brain-Inspired Large Language Models for Long-Context Understanding

Kangcong Li
Peng Ye
Chongjun Tu
Lin Zhang
Chunfeng Song
Jiamin Wu
Tao Yang
Qihao Zheng

While Large Language Models (LLMs) demonstrate strong performance across domains, their long-context capabilities are limited by transient neural activations causing information decay and unstructured feed-forward network (FFN) weights leading to semantic fragmentation. Inspired by the brain’s working memory and cortical modularity, we propose PaceLLM, featuring two innovations: (1) a Persistent Activity (PA) Mechanism that mimics prefrontal cortex (PFC) neurons’ persistent firing by introducing an activation-level memory bank to dynamically retrieve, reuse, and update critical FFN states, addressing contextual decay; and (2) Cortical Expert (CE) Clustering that emulates task-adaptive neural specialization to reorganize FFN weights into semantic modules, establishing cross-token dependencies and mitigating fragmentation. Extensive evaluations show that PaceLLM achieves 6% improvement on LongBench’s Multi-document QA and 12. 5–17. 5% performance gains on $\infty$-Bench tasks, while extending measurable context length to 200K tokens in Needle-In-A-Haystack (NIAH) tests. This work pioneers brain-inspired LLM optimization and is complementary to other works. Besides, it can be generalized to any model and enhance their long-context performance and interpretability without structural overhauls.

PDF Details

AAAI Conference 2025 Conference Paper

Representing Sounds as Neural Amplitude Fields: A Benchmark of Coordinate-MLPs and a Fourier Kolmogorov-Arnold Framework

Linfei Li
Lin Zhang
Zhong Wang
Fengyi Zhang
Zelin Li
Ying Shen

Although Coordinate-MLP-based implicit neural representations have excelled in representing radiance fields, 3D shapes, and images, their application to audio signals remains underexplored. To fill this gap, we investigate existing implicit neural representations, from which we extract 3 types of positional encoding and 16 commonly used activation functions. Through combinatorial design, we establish the first benchmark for Coordinate-MLPs in audio signal representations. Our benchmark reveals that Coordinate-MLPs require complex hyperparameter tuning and frequency-dependent initialization, limiting their robustness. To address these issues, we propose Fourier-ASR, a novel framework based on the Fourier series theorem and the Kolmogorov-Arnold representation theorem. Fourier-ASR introduces Fourier Kolmogorov-Arnold Networks (Fourier-KAN), which leverage periodicity and strong nonlinearity to represent audio signals, eliminating the need for additional positional encoding. Furthermore, a Frequency-adaptive Learning Strategy (FaLS) is proposed to enhance the convergence of Fourier-KAN by capturing high-frequency components and preventing overfitting of low-frequency signals. Extensive experiments conducted on natural speech and music datasets reveal that: (1) well-designed positional encoding and activation functions in Coordinate-MLPs can effectively improve audio representation quality; and (2) Fourier-ASR can robustly represent complex audio signals without extensive hyperparameter tuning. Looking ahead, the continuity and infinite resolution of implicit audio representations make our research highly promising for tasks such as audio compression, synthesis, and generation.

PDF Details DOI

TIST Journal 2025 Journal Article

The Evaluation Framework and Benchmark for Large Language Models in the Government Affairs Domain

Shuo Liu
Lin Zhang
Weidong Liu
Jianfeng Zhang
Donghui Gao
Xiaofeng Jia

The rapid evolution of AI has driven advancements across numerous sectors. In the domain of government affairs, large language models (LLMs) hold significant potential for applications such as policy analysis, data processing, and decision support. However, their adoption in government settings faces considerable challenges, including data accessibility issues, the absence of standardized evaluation criteria, and concerns regarding model accuracy, reliability, and security. To address these challenges, we propose a comprehensive evaluation framework specifically designed for LLMs in government affairs. Built on modular principles, this framework ensures adaptability across various industries. Additionally, we introduce the Multi-Scenario Government Affairs Benchmark (MSGABench 1 ) dataset, a Chinese-language dataset specifically crafted to meet the practical needs of government professionals. Employing the proposed framework and the MSGA dataset, we conducted an empirical evaluation of 15 prominent LLMs, revealing critical insights: (1) Performance: Many models demonstrated low accuracy and reliability, particularly under minor input variations, with some dropping below 35% accuracy, whereas GPT-4 achieved above 95% reliability; (2) Security and Compliance: Significant concerns were identified, including privacy vulnerabilities, legal compliance risks, and persistent biases, which may hinder secure deployments in government contexts; (3) Task Avoidance: Certain models exhibited excessive caution, often avoiding responses to basic tasks like document classification and government-related inquiries, which restricts their usability. These findings highlight essential limitations and opportunities for improvement, contributing to the safe and effective application of LLMs in the government sector.

Details DOI

AAAI Conference 2025 Conference Paper

Towards Audio-Visual Navigation in Noisy Environments: A Large-Scale Benchmark Dataset and an Architecture Considering Multiple Sound-Sources

Zhanbo Shi
Lin Zhang
Linfei Li
Ying Shen

Audio-visual navigation has received considerable attention in recent years. However, the majority of related investigations have focused on single sound-source scenarios. Studies in this field for multiple sound-source scenarios remain underexplored due to the limitations of two aspects. First, the existing audio-visual navigation dataset only has limited audio samples, making it difficult to simulate diverse multiple sound-source environments. Second, existing navigation frameworks are mainly designed for single sound-source scenarios, thus their performance is severely reduced in multiple sound-source scenarios. In this work, we make an attempt to fill in these two research gaps to some extent. First, we establish a large-scale BEnchmark Dataset for Audio-Vsual Navigation, namely BeDAViN. This dataset consists of 2,258 audio samples with a total duration of 10.8 hours, which is more than 33 times longer than the existing audio dataset employed in the audio-visual navigation task. Second, we propose a new Embodied Navigation framework for MUltiple Sound-Sources Scenarios called ENMuS3. There are mainly two essential components in ENMuS3, the sound event descriptor and the multi-scale scene memory transformer. The former component equips the agent with the ability to extract spatial and semantic features of the target sound-source among multiple sound-sources, while the latter provides the ability to track the target object effectively in noisy environments. Experimental results on our BeDAViN show that ENMuS3 strongly outperforms its counterparts with a significant improvement in success rates across diverse scenarios.

PDF Details DOI

NeurIPS Conference 2024 Conference Paper

3DET-Mamba: Causal Sequence Modelling for End-to-End 3D Object Detection

Mingsheng Li
Jiakang Yuan
Sijin Chen
Lin Zhang
Anyu Zhu
Xin Chen
Tao Chen

Transformer-based architectures have been proven successful in detecting 3D objects from point clouds. However, the quadratic complexity of the attention mechanism struggles to encode rich information as point cloud resolution increases. Recently, state space models (SSM) such as Mamba have gained great attention due to their linear complexity and long sequence modeling ability for language understanding. To exploit the potential of Mamba on 3D scene-level perception, for the first time, we propose 3DET-Mamba, which is a novel SSM-based model designed for indoor 3d object detection. Specifically, we divide the point cloud into different patches and use a lightweight yet effective Inner Mamba to capture local geometric information. To observe the scene from a global perspective, we introduce a novel Dual Mamba module that models the point cloud in terms of spatial distribution and continuity. Additionally, we design a Query-aware Mamba module that decodes context features into object sets under the guidance of learnable queries. Extensive experiments demonstrate that 3DET-Mamba surpasses previous 3DETR on indoor 3D detection benchmarks such as ScanNet, improving AP25/AP50 from 65. 0\%/47. 0\% to 70. 4\%/54. 4\%, respectively.

PDF Details DOI

EAAI Journal 2024 Journal Article

Reinforcement learning-based distributed impedance control of robots for compliant operation in tight interaction tasks

Pengjie Xu
Zhenyu Li
Xun Liu
Tianrui Zhao
Lin Zhang
Yanzheng Zhao

It is challenging to achieve compliant operation in tight interaction tasks, where closed-loop constraints are formed between the robots or between a robot and the environment. Complex dynamic interactions and uncertain parameters degrade the performance of model-based controllers. In this paper, a reinforcement learning-based distributed impedance control approach is proposed for these tight interaction tasks. Two aspects are considered to ensure compliant operation for the robots. First, a distributed impedance model is established through the design of reasonable independence states and nodes networks. Second, the reinforcement learning agent is designed to make decision for adjusting impedance parameters. The trained parameters are integrated into the designed impedance model, and a model-based controller is then employed for compliant control for the robots. The effectiveness is validated under two different tight interaction scenarios via co-simulation. Compliant operation can be achieved whether between robots or between a robot and the environment.

Details DOI

AAAI Conference 2024 Conference Paper

Xiezhi: An Ever-Updating Benchmark for Holistic Domain Knowledge Evaluation

Zhouhong Gu
Xiaoxuan Zhu
Haoning Ye
Lin Zhang
Jianchen Wang
Yixin Zhu
Sihang Jiang
Zhuozhi Xiong

New Natural Langauge Process~(NLP) benchmarks are urgently needed to align with the rapid development of large language models (LLMs). We present Xiezhi, the most comprehensive evaluation suite designed to assess holistic domain knowledge.Xiezhi comprises multiple-choice questions across 516 diverse disciplines ranging from 13 different subjects with 249,587 questions and accompanied by Xiezhi-Specialty with 14,041 questions and Xiezhi-Interdiscipline with 10,746 questions. We conduct evaluation of the 47 cutting-edge LLMs on Xiezhi. Results indicate that LLMs exceed average performance of humans in science, engineering, agronomy, medicine, and art, but fall short in economics, jurisprudence, pedagogy, literature, history, and management. All the evaluation code and data are open sourced in https://github.com/MikeGu721/XiezhiBenchmark

PDF Details DOI

ICML Conference 2023 Conference Paper

D2Match: Leveraging Deep Learning and Degeneracy for Subgraph Matching

Xuanzhou Liu
Lin Zhang
Jiaqi Sun
Yujiu Yang 0001
Haiqin Yang

Subgraph matching is a fundamental building block for graph-based applications and is challenging due to its high-order combinatorial nature. Existing studies usually tackle it by combinatorial optimization or learning-based methods. However, they suffer from exponential computational costs or searching the matching without theoretical guarantees. In this paper, we develop $D^2$Match by leveraging the efficiency of Deep learning and Degeneracy for subgraph matching. More specifically, we first prove that subgraph matching can degenerate to subtree matching, and subsequently is equivalent to finding a perfect matching on a bipartite graph. We can then yield an implementation of linear time complexity by the built-in tree-structured aggregation mechanism on graph neural networks. Moreover, circle structures and node attributes can be easily incorporated in $D^2$Match to boost the matching performance. Finally, we conduct extensive experiments to show the superior performance of our $D^2$Match and confirm that our $D^2$Match indeed exploits the subtrees and differs from existing GNNs-based subgraph matching methods that depend on memorizing the data distribution divergence.

Details

ICML Conference 2023 Conference Paper

Do Not Train It: A Linear Neural Architecture Search of Graph Neural Networks

Peng Xu 0052
Lin Zhang
Xuanzhou Liu
Jiaqi Sun
Yue Zhao 0016
Haiqin Yang
Bei Yu 0001

Neural architecture search (NAS) for Graph neural networks (GNNs), called NAS-GNNs, has achieved significant performance over manually designed GNN architectures. However, these methods inherit issues from the conventional NAS methods, such as high computational cost and optimization difficulty. More importantly, previous NAS methods have ignored the uniqueness of GNNs, where GNNs possess expressive power without training. With the randomly-initialized weights, we can then seek the optimal architecture parameters via the sparse coding objective and derive a novel NAS-GNNs method, namely neural architecture coding (NAC). Consequently, our NAC holds a no-update scheme on GNNs and can efficiently compute in linear time. Empirical evaluations on multiple GNN benchmark datasets demonstrate that our approach leads to state-of-the-art performance, which is up to $200\times$ faster and $18. 8%$ more accurate than the strong baselines.

Details

ICML Conference 2023 Conference Paper

Feature Expansion for Graph Neural Networks

Jiaqi Sun
Lin Zhang
Guangyi Chen 0002
Peng Xu
Kun Zhang 0001
Yujiu Yang 0001

Graph neural networks aim to learn representations for graph-structured data and show impressive performance in node classification. Recently, many methods have studied the representations of GNNs from the perspective of optimization goals and spectral graph theory. However, the feature space that dominates representation learning has not been systematically studied in graph neural networks. In this paper, we propose to fill this gap by analyzing the feature space of both spatial and spectral models. We decompose graph neural networks into determined feature spaces and trainable weights, providing the convenience of studying the feature space explicitly using matrix space analysis. In particular, we find theoretically that the feature space tends to be linearly correlated due to repeated aggregations. In this case, the feature space is bounded by the poor representation of shared weights or the limited dimensionality of node attributes in existing models, leading to poor performance. Motivated by these findings, we propose 1) feature subspaces flattening and 2) structural principal components to expand the feature space. Extensive experiments verify the effectiveness of our proposed more comprehensive feature space, with comparable inference time to the baseline, and demonstrate its efficient convergence capability.

Details

AAAI Conference 2023 Conference Paper

Self-Supervised Bidirectional Learning for Graph Matching

Wenqi Guo
Lin Zhang
Shikui Tu
Lei Xu

Deep learning methods have demonstrated promising performance on the NP-hard Graph Matching (GM) problems. However, the state-of-the-art methods usually require the ground-truth labels, which may take extensive human efforts or be impractical to collect. In this paper, we present a robust self-supervised bidirectional learning method (IA-SSGM) to tackle GM in an unsupervised manner. It involves an affinity learning component and a classic GM solver. Specifically, we adopt the Hungarian solver to generate pseudo correspondence labels for the simple probabilistic relaxation of the affinity matrix. In addition, a bidirectional recycling consistency module is proposed to generate pseudo samples by recycling the pseudo correspondence back to permute the input. It imposes a consistency constraint between the pseudo affinity and the original one, which is theoretically supported to help reduce the matching error. Our method further develops a graph contrastive learning jointly with the affinity learning to enhance its robustness against the noise and outliers in real applications. Experiments deliver superior performance over the previous state-of-the-arts on five real-world benchmarks, especially under the more difficult outlier scenarios, demon- strating the effectiveness of our method.

PDF Details DOI

ECAI Conference 2023 Conference Paper

Specializing Small Language Models Towards Complex Style Transfer via Latent Attribute Pre-Training

Yongfeng Huang 0001
Xin Chen
Lin Zhang

In this work, we introduce the concept of complex text style transfer tasks, and constructed complex text datasets based on two widely applicable scenarios. Our dataset is the first large-scale data set of its kind, with 700 rephrased sentences and 1, 000 sentences from the game Genshin Impact. While large language models (LLM) have shown promise in complex text style transfer, they have drawbacks such as data privacy concerns, network instability, and high deployment costs. To address these issues, we explore the effectiveness of small models (less than T5-3B) with implicit style pre-training through contrastive learning. We also propose a method for automated evaluation of text generation quality based on alignment with human evaluations using ChatGPT. Finally, we compare our approach with existing methods and show that our model achieves state-of-art performances of few-shot text style transfer models.

Details

JBHI Journal 2022 Journal Article

BDBB: A Novel Beta-Distribution-Based Biclustering Algorithm for Revealing Local Co-Methylation Patterns in Epi-Transcriptome Profiling Data

Zhaoyang Liu
Yuteng Xiao
Hongsheng Yin
Xiaodan Li
Shutao Chen
Kaijian Xia
Lin Zhang

N6-methyladenosine (m 6 A) has been shown to play crucial roles in RNA metabolism, physiology, and pathological processes. However, the specific regulatory mechanisms of most methylation sites remain uncharted due to the complexity of life processes. Biological experimental methods are costly to solve this problem, and computational methods are relatively lacking. The discovery of local co-methylation patterns (LCPs) of m 6 A epi-transcriptome data can benefit to solve the above problems. Based on this, we propose a novel biclustering algorithm based on the beta distribution (BDBB), which realizes the mining of LCPs of m 6 A epi-transcriptome data. BDBB employs the Gibbs sampling method to complete parameter estimation. In the process of modeling, LCPs are recognized as sharp beta distributions compared to the background distribution. Simulation study showed BDBB can extract all the three actual LCPs implanted in the background data and the overlap conditions between them with considerable accuracy (almost close to 100%). On MeRIP-Seq data of 69, 446 methylation sites under 32 experimental conditions from 10 human cell lines, BDBB unveiled two LCPs, and Gene Ontology (GO) enrichment analysis showed that they were enriched in histone modification and embryo development, etc. important biological processes respectively. The GOE_Score scoring indicated that the biclustering results of BDBB in the m 6 A epi-transcriptome data are more biologically meaningful than the results of other biclustering algorithms.

Details DOI

ICRA Conference 2022 Conference Paper

CCRobot-V: A Silkworm-Like Cooperative Cable-Climbing Robotic System for Cable Inspection and Maintenance

Zhenliang Zheng
Ning Ding 0003
Huaping Chen 0005
Xiaoli Hu
Zhihao Zhu
Xueqi Fu
Wenchao Zhang
Lin Zhang

This paper presents CCRobot-V, the fifth version of CCRobot, a cooperative serial multi-robot system for bridge cable inspection and maintenance that uses silkworm-like locomotion to climb the entire length of super-long stay cable at high speeds while carrying heavy inspection/maintenance equipment. CCRobot-V consists of one climbing precursor robot, one inspection/maintenance robot, several cable-carrying robots, and a power-tethered cable guiding system. The pre-cursor robot is the “head, ” which leads the affiliated sub-robots along the bridge cable. Every sub-robot possesses a pair of self-locking palms. When a sub-robot grips on the bridge cable with its palms, it becomes a fixed anchor point that allows the adjacent sub-robots in front and back to use winches and steel wires to pull themselves upward. With this cooperative multi-robot system, cable inspection/maintenance tasks can be divided into several functional units, with each inspection/maintenance equipment installed separately on a customized sub-robot. Thus, CCRobot-V provides a complete mobile inspection/maintenance work line for a bridge cable. The experimental and field tests demonstrate CCRobot-V's high climbing speed, high payload capacity, and full-length cable moving capability. It has the potential application value for the actual bridge cable inspection/maintenance.

Details

NeurIPS Conference 2022 Conference Paper

Exact Shape Correspondence via 2D graph convolution

Barakeel Fanseu Kamhoua
Lin Zhang
Yongqiang Chen
Han Yang
MA KAILI
Bo Han
Bo Li
James Cheng

For exact 3D shape correspondence (matching or alignment), i. e. , the task of matching each point on a shape to its exact corresponding point on the other shape (or to be more specific, matching at geodesic error 0), most existing methods do not perform well due to two main problems. First, on nearly-isometric shapes (i. e. , low noise levels), most existing methods use the eigen-vectors (eigen-functions) of the Laplace Beltrami Operator (LBO) or other shape descriptors to update an initialized correspondence which is not exact, leading to an accumulation of update errors. Thus, though the final correspondence may generally be smooth, it is generally inexact. Second, on non-isometric shapes (noisy shapes), existing methods are generally not robust to noise as they usually assume near-isometry. In addition, existing methods that attempt to address the non-isometric shape problem (e. g. , GRAMPA) are generally computationally expensive and do not generalise to nearly-isometric shapes. To address these two problems, we propose a 2D graph convolution-based framework called 2D-GEM. 2D-GEM is robust to noise on non-isometric shapes and with a few additional constraints, it also addresses the errors in the update on nearly-isometric shapes. We demonstrate the effectiveness of 2D-GEM by achieving a high accuracy of 90. 5$\%$ at geodesic error 0 on the non-isometric benchmark SHREC16, i. e. , TOPKIDS (while being much faster than GRAMPA), and on nearly-isometric benchmarks by achieving a high accuracy of 92. 5$\%$ on TOSCA and 84. 9$\%$ on SCAPE at geodesic error 0.

PDF Details

TIST Journal 2022 Journal Article

Intrinsic Performance Influence-based Participant Contribution Estimation for Horizontal Federated Learning

Lin Zhang
Lixin Fan
Yong Luo
Ling-Yu Duan

The rapid development of modern artificial intelligence technique is mainly attributed to sufficient and high-quality data. However, in the data collection, personal privacy is at risk of being leaked. This issue can be addressed by federated learning, which is proposed to achieve efficient model training among multiple data providers without direct data access and aggregation. To encourage more parties owning high-quality data to participate in the federated learning, it is important to evaluate and reward the participant contribution in a reasonable, robust, and efficient manner. To achieve this goal, we propose a novel contribution estimation method: Intrinsic Performance Influence-based Contribution Estimation (IPICE). In particular, the class-level intrinsic performance influence is adopted as the contribution estimation criteria in IPICE, and a neural network is employed to exploit the non-linear relationship between the performance change and estimated contribution. Extensive experiments are conducted on various datasets, and the results demonstrate that IPICE is more accurate and stable than the counterpart in various data distribution settings. The computational complexity is significantly reduced in our IPICE, especially when a new party joins the federation. IPICE assigns small contributions to bad/garbage data and thus prevent them from participating and deteriorating the learning ecosystem.

Details DOI

EAAI Journal 2022 Journal Article

Motion parameters measurement of user-defined key points using 3D pose estimation

Xin Wu
Yonghui Wang
Lei Chen
Lin Zhang
Lianming Wang

Motion parameters measurement is essential for understanding animal behavior, exploring the laws of object motion, and studying control methods. Nowadays, advanced computer vision based on machine learning technology supports markerless object tracking in 2D videos. However, due to the fact that all objects move in three-dimensional space, this paper introduces a method of measuring motion parameters using 3D pose estimation. First, an enhanced iterative bundle adjustment algorithm is proposed for multi-camera calibration in a multi-camera vision system by adding two control parameters, which dramatically reduces the reprojection error of multi-camera calibration and lays the foundation for high-precision triangulation. Then, a new spatiotemporal loss function is proposed, which considers the relationship between key points that do not constitute limbs, thereby improving triangulation accuracy. The new multi-camera calibration algorithm is evaluated on ChArUco and 3D pose estimation for metronome, planet pendulum, human hand, Koi, and cheetah. The experimental results show that: (1) the two hyper-parameters in the enhanced iterative bundle adjustment algorithm effectively suppress the influence of noise and play a good role in reducing the reprojection error of multi-camera calibration; (2) the spatiotemporal loss function has a strong constraining ability, the time loss can stabilize high frame rate video triangulation to maintain accuracy, while the space loss can improve the accuracy of triangulation for more complex structures; (3) multi-view data fusion is also conducive to improving the accuracy of triangulation. Moreover, the method was successfully applied to some actual measurement scenes: (1) the accurate measurement of the frequency of a metronome; and (2) the success measurement of the movement of a Koi, which conforms to the basic model of fish swimming. Some dynamic measurement results are displayed at https: //github. com/wux024/AdamPose.

Details DOI

YNICL Journal 2020 Journal Article

Cerebral blood flow characteristics following hemodialysis initiation in older adults: A prospective longitudinal pilot study using arterial spin labeling imaging

Xiufeng Li
Yelena X. Slinin
Lin Zhang
Donald R. Dengel
David Tupper
Gregory J. Metzger
Anne M. Murray

PURPOSE: To investigate cerebral blood flow (CBF) characteristics before and after hemodialysis initiation and their longitudinal associations with global cognitive function in older adults. METHODS: A cohort of 17 older end-stage renal disease patients anticipating standard thrice-weekly hemodialysis and a group of 11 age- and sex-matched healthy control volunteers were recruited for brain perfusion imaging studies using arterial spin labeling. Hemodialysis patients participated in a prospective longitudinal study using brain magnetic resonance imaging and global cognitive assessment using the Modified Mini-Mental State Examination (3MS) at two time points: baseline, 2.9 ± 0.9 months before, and follow-up, 6.4 ± 2.4 months after hemodialysis initiation. Healthy controls were imaged once using the same protocol. CBF analyses were performed globally in grey and white matter and regionally in the hippocampus and orbitofrontal cortex. Covariate-adjusted linear mixed-effects models were used for statistical analyses (significance: p < 0.05; marginal significance: p < 0.1). RESULTS: At baseline, global and regional CBF was significantly higher in hemodialysis patients than in healthy controls. However, after approximately 6 months of hemodialysis, CBF declined substantially in hemodialysis patients, and became comparable to those in healthy controls. Specifically, in the hemodialysis patients, CBF declined non-significantly globally for grey and white matter and significantly regionally in the hippocampus and orbitofrontal cortex. Marginally significant associations were observed between 3MS scores and regional CBF measurements in the hippocampus and orbitofrontal cortex at baseline and follow-up, and between longitudinal changes. CONCLUSION: The significant decline in CBF after hemodialysis initiation and the observed association between longitudinal changes in regional CBF and 3MS scores suggest that decreased brain perfusion may contribute to the observed cognitive decline.

Details DOI

AAAI Conference 2019 Conference Paper

An Efficient Approach to Informative Feature Extraction from Multimodal Data

Lichen Wang
Jiaxiang Wu
Shao-Lun Huang
Lizhong Zheng
Xiangxiang Xu
Lin Zhang
Junzhou Huang

One primary focus in multimodal feature extraction is to find the representations of individual modalities that are maximally correlated. As a well-known measure of dependence, the Hirschfeld-Gebelein-Rényi (HGR) maximal correlation becomes an appealing objective because of its operational meaning and desirable properties. However, the strict whitening constraints formalized in the HGR maximal correlation limit its application. To address this problem, this paper proposes Soft-HGR, a novel framework to extract informative features from multiple data modalities. Specifically, our framework prevents the “hard” whitening constraints, while simultaneously preserving the same feature geometry as in the HGR maximal correlation. The objective of Soft-HGR is straightforward, only involving two inner products, which guarantees the efficiency and stability in optimization. We further generalize the framework to handle more than two modalities and missing modalities. When labels are partially available, we enhance the discriminative power of the feature representations by making a semi-supervised adaptation. Empirical evaluation implies that our approach learns more informative feature mappings and is more efficient to optimize.

PDF Details

IS Journal 2017 Journal Article

Learning Geographical and Mobility Factors for Mobile Application Recommendation

Konglin Zhu
Lin Zhang
Achille Pattavina

With myriad features and functionalities, mobile app users have the option to run different types of apps when they move to different locations. For a specific place, the decision process involved in choosing a mobile app can be complex and influenced by various factors, such as app popularity, user preferences, geographical influences, and user mobility behaviors. Although several researchers have studied recommendation in mobile apps, they omitted an integrated analysis of the joint effect of multiple factors from a geographical perspective. This article proposes a novel location-based probabilistic factor analysis mechanism that considers multiple factors to help people visiting a new location get an appropriate mobile app recommendation. In particular, the authors model mobile app usage from a geographical perspective. Experimental results on real mobile usage data show that the proposed recommendation method outperforms baseline algorithms by 30 percent.

Details DOI

YNIMG Journal 2016 Journal Article

Dissociable roles of internal feelings and face recognition ability in facial expression decoding

Lin Zhang
Yiying Song
Ling Liu
Jia Liu

The problem of emotion recognition has been tackled by researchers in both affective computing and cognitive neuroscience. While affective computing relies on analyzing visual features from facial expressions, it has been proposed that humans recognize emotions by internally simulating the emotional states conveyed by others' expressions, in addition to perceptual analysis of facial features. Here we investigated whether and how our internal feelings contributed to the ability to decode facial expressions. In two independent large samples of participants, we observed that individuals who generally experienced richer internal feelings exhibited a higher ability to decode facial expressions, and the contribution of internal feelings was independent of face recognition ability. Further, using voxel-based morphometry, we found that the gray matter volume (GMV) of bilateral superior temporal sulcus (STS) and the right inferior parietal lobule was associated with facial expression decoding through the mediating effect of internal feelings, while the GMV of bilateral STS, precuneus, and the right central opercular cortex contributed to facial expression decoding through the mediating effect of face recognition ability. In addition, the clusters in bilateral STS involved in the two components were neighboring yet separate. Our results may provide clues about the mechanism by which internal feelings, in addition to face recognition ability, serve as an important instrument for humans in facial expression decoding.

Details DOI

ICAPS Conference 2004 Conference Paper

Decision-Theoretic Military Operations Planning

Douglas Aberdeen
Sylvie Thiébaux
Lin Zhang

Military operations planning involves concurrent actions, resource assignment, and conflicting costs. Individual tasks sometimes fail with a known probability, promoting a decision-theoretic approach. The planner must choose between multiple tasks that achieve similar outcomes but have different costs. The military domain is particularly suited to automated methods because hundreds of tasks, specified by many planning staff, need to be quickly and robustly coordinated. The authors are not aware of any previous planners that handle all characteristics of the operations planning domain in a single package. This paper shows that problems with such features can be successfully approached by realtime heuristic search algorithms, operating on a formulation of the problem as a Markov decision process. Novel automatically generated heuristics, and classic caching methods, allow problems of interesting sizes to be handled. Results are presented on data provided by the Australian Defence Science and Technology Organisation.

Details

YNIMG Journal 2000 Journal Article

Magnetic Resonance Microscopy of the C57BL Mouse Brain

Helene Benveniste
Katie Kim
Lin Zhang
G.Allan Johnson

With the rapid progression in gene technologies, transgenic, targeted, and chemically induced mutations in mice are continually created. The major goal of these studies is to understand and characterize the effects of genotype on anatomy, physiology, and behavior and ultimately the role of genotype in development of disease. The demand for imaging techniques with high spatial resolution potential is rising because such imaging tools would expedite anatomical phenotyping in the genetically altered mice. Magnetic resonance microscopy (MRM) is a noninvasive, inherently three-dimensional (3D) imaging technique capable of visualizing several anatomical structures in the small mouse. The 3D nature of MRM also allows for interpretation of complex spatial relationships between substructures, which is important when phenotyping anatomically. The goal of this paper is to systematically describe three major brain regions in the C57BL/6J mouse at microanatomical spatial resolution ranges using in vitro MRM. We explore different MR contrast parameters, voxel sizes, and signal-to-noise ratios to best characterize C57BL/6J mouse brain microstructure by MRM. Further, we compare all MRM images with Nissl-stained brain sections. Major findings were as follows: T2* MR images visualized several gross anatomical regions in the mouse brain but not, for example, subregions within the hippocampus. Diffusion proton stains on the other hand were superior to T2* MR images and delineated many subregions within the hippocampus proper. Finally, contrast enhancement facilitated visualization of hippocampal anatomy on the T2* MR images. The results of this study are part of an ongoing initiative at our Center focused on creating a complete C57BL/6J mouse anatomical 3D image database by MRM.

Details DOI