Searcharxiv⌕ Search

arXiv subjects

Jinlong Hu

Publications and source records attributed to Jinlong Hu.

At least 19 recordsLinked to original sources

DGPExplainer: Generating Prototype Sequence Explanations for Dynamic Graph Neural Networks

Graph Neural Networks (GNNs) have been successfully applied to learning from graph-structured data. However, applying them to dynamic graphs-where nodes and edges evolve over time-presents significant challenges, particularly regarding explainability. While explanation techniques for static GNNs are well established, they do not readily extend to dynamic GNNs (DGNNs) due to the added complexity introduced by temporal dynamics and spatiotemporal dependencies, especially when generating model-level explanations. To address this gap, we introduce DGPExplainer, a post-hoc, prototype-based explainer specifically designed for DGNN models that transforms their complex decision-making processes into human-understandable, analogical, case-based explanations. DGPExplainer first employs local explanation methods to identify important elements within temporal snapshots of dynamic graph sequences. It then decomposes the graph at each time step into distinct prototypes, thereby symbolizing the dynamic graph sequence. Finally, we design a sequence generation method that uses these symbolized sequences to learn and generate class-specific dynamic graph pattern sequences that serve as prototype explanations. Experiments on both synthetic and real-world datasets demonstrate that our method effectively produces prototype-based explanations for DGNN models across multiple classes.

cs.LG↗

HydroTuring: A physics probe benchmark testing mass, energy, and momentum conservation in hydrologic models

Hydrologic models, including deep learning systems now deployed for operational flood forecasting, are evaluated primarily by how closely their simulated hydrographs match observed streamflow. However, high hydrograph skill does not guarantee physical consistency: in a synthetic calibration experiment, a conceptual model that loses 15% of its water input to an unreported sink still reaches a Nash-Sutcliffe efficiency of 0.93, and no shared benchmark tests whether models obey fundamental conservation laws. Here we introduce HydroTuring (v0.1.0), a black-box benchmark of 37 physics probes that test mass (24 probes), energy (7) and momentum (6) conservation in hydrologic models. Each probe runs a controlled experiment on forcing generated at run time from a recorded seed and applies binary pass or fail criteria validated against 18 physically exact and 59 deliberately broken reference models. Across 32 evaluated models (3 exact references, 7 AI-based models and 22 process-based models), only the 3 exact references pass every probe among the 28 models scored on 5 or more probes. Process-based and AI-based models pass similar shares of their scored probes (76% and 74%), and a pre-trained flood-forecasting LSTM fails 6 of the 11 runoff-only probes it can be scored on. Furthermore, 49% of all 1,184 model-probe pairs remain unscored because models omit the required state or flux variables, while the scored failures occur in perturbation responses and state bounds rather than long-term water-budget closure. HydroTuring provides a reproducible smoke test for model development, peer review and operational deployment under an open community protocol.

physics.geo-ph↗

SetOPD: From Few Visual Exemplars to Multimodal Candidate Sets for Remote-Sensing Open-Prompt Detection

Open-prompt detectors allow users to specify targets with text, visual exemplars, or both. We argue that existing designs underuse the visual modality in two ways. First, multiple exemplars are commonly compressed into a single class-level embedding. This textualizes visual prompting: the resulting vector plays the role of another class name, is often aligned to or injected into the text pathway, and may be suboptimal when only a few heterogeneous exemplars are available. Second, existing methods interact primarily in prompt or representation space, before modality-specific detection states are formed. We address both issues from a set perspective: \setopd preserves modality-specific decoding states from a shared prompt-conditioned initialization and performs explicit multimodal collaboration at the candidate-state level. For the first issue, we introduce \br prompting, which reads every boxed exemplar in its full scene context and decomposes the pooled support evidence into a Base anchor and a learned Residual correction; the resulting prompt has fixed capacity regardless of the number of examples and drives its own visual detection pathway. For the second, we recast text--visual collaboration from representation-level fusion into a candidate-set modeling problem. Paired-Query Arbitration (\pqa) then performs explicit cross-modal state arbitration only after modality-specific candidate states have been formed. The two readers share query initialization so their candidates are paired by index; a learned gate arbitrates within each pair, followed by a permutation-equivariant module that reasons over the fused set.

cs.CV↗

Hi-OPD: Hierarchy-Aware Open-Prompt Detection for Remote Sensing Images

Hi-OPD addresses a failure mode left uncontrolled by flat open-prompt training: descendant retrieval need not persist under ancestor queries when multi-source remote sensing annotations exhibit inconsistent granularity and missing labels. A detector may localize \textit{car} and \textit{van} under atomic prompts yet miss the same instances under \textit{vehicle}; flat AP does not expose this cross-level inconsistency. We propose Hi-OPD, a hierarchy-aware open-prompt detector, and construct RS153-HierOPD from 175,644 retained training image/tile records and 3.48M boxes mapped to 153 atomic categories with sparse hierarchy and alias relations. Hi-OPD learns ancestor retrieval through hierarchy-safe negative sampling, path multi-positive supervision, and one-way upward consistency, while per-source risk exclusion handles potentially missing labels. ConvVPE converts K-shot support boxes into text-compatible embeddings using detector-native features and the shared contrastive head. On Track A, Hi-OPD obtains 79.7/72.3 AP50 on DIOR/DOTA-v2.0, above the literature-reported OpenRSD results of 76.7/71.8. Under controlled training on the original converted annotations, the full hierarchy recipe raises DOTA-v2.0 parent AP50 from 7.2 to 71.5 and FAIR1M grandparent AP50 from 31.6 to 71.4, while DOTA-v2.0 atomic AP50 changes from 71.4 to 72.3. The text path reaches 99.7% CAR50 (0.3% violation) across the three common sources and 99.9%/0.1% on FAIR1M grandparent relations. On held-out VEDAI, text AP50 is 75.9, 6.2 points above OpenRSD. Joint AP and CAR show that explicit hierarchy training repairs this failure mode while retaining atomic detection and prompt transfer.

cs.CV↗

Fusion Learning from Dynamic Functional Connectivity: Combining the Amplitude and Phase of fMRI Signals to Identify Brain Disorders

Dynamic functional connectivity (dFC) derived from resting-state functional magnetic resonance imaging (fMRI) has been extensively utilized in brain science research. The sliding window correlation (SWC) method is a widely used approach for constructing dFC by computing correlation coefficients between amplitude time series of signals from pairs of brain regions. In this study, we propose an integrated approach that incorporates both amplitude and phase information of fMRI signals to improve the detection of brain disorders. Specifically, we introduce a multi-scale fusion learning framework, namely MSFL, which leverages two complementary dFC features derived from SWC and phase synchronization (PS). Here, SWC captures amplitude correlations, while PS measures phase coherence within dFC. We evaluated the efficacy of MSFL in classifying autism spectrum disorder and major depressive disorder using two publicly available datasets: ABIDE I and REST-meta-MDD, respectively. The results indicate that MSFL significantly outperforms existing comparative models. Moreover, we performed model explanation analysis using the SHAP framework, which showed that both types of dFC features from SWC and PS contribute to detecting brain disorders.

q-bio.NC↗

GCFX: Generative Counterfactual Explanations for Deep Graph Models at the Model Level

Deep graph learning models have demonstrated remarkable capabilities in processing graph-structured data and have been widely applied across various fields. However, their complex internal architectures and lack of transparency make it difficult to explain their decisions, resulting in opaque models that users find hard to understand and trust. In this paper, we explore model-level explanation techniques for deep graph learning models, aiming to provide users with a comprehensive understanding of the models' overall decision-making processes and underlying mechanisms. Specifically, we address the problem of counterfactual explanations for deep graph learning models by introducing a generative model-level counterfactual explanation approach called GCFX, which is based on deep graph generation. This approach generates a set of high-quality counterfactual explanations that reflect the model's global predictive behavior by leveraging an enhanced deep graph generation framework and a global summarization algorithm. GCFX features an architecture that combines dual encoders, structure-aware taggers, and Message Passing Neural Network decoders, enabling it to accurately learn the true latent distribution of input data and generate high-quality, closely related counterfactual examples. Subsequently, a global counterfactual summarization algorithm selects the most representative and comprehensive explanations from numerous candidate counterfactuals, providing broad insights into the model's global predictive patterns. Experiments on a synthetic dataset and several real-world datasets demonstrate that GCFX outperforms existing methods in terms of counterfactual validity and coverage while maintaining low explanation costs, thereby offering crucial support for enhancing the practicality and trustworthiness of global counterfactual explanations.

cs.LG↗

WebRouter: Query-specific Router via Variational Information Bottleneck for Cost-sensitive Web Agent

LLM-brained web agents offer powerful capabilities for web automation but face a critical cost-performance trade-off. The challenge is amplified by web agents' inherently complex prompts that include goals, action histories, and environmental states, leading to degraded LLM ensemble performance. To address this, we introduce WebRouter, a novel query-specific router trained from an information-theoretic perspective. Our core contribution is a cost-aware Variational Information Bottleneck (ca-VIB) objective, which learns a compressed representation of the input prompt while explicitly penalizing the expected operational cost. Experiments on five real-world websites from the WebVoyager benchmark show that WebRouter reduces operational costs by a striking 87.8\% compared to a GPT-4o baseline, while incurring only a 3.8\% accuracy drop.

cs.CL↗

Runtime-optimized Multi-way Stream Join Operator for Large-scale Streaming data

Streaming computing enables the real-time processing of large volumes of data and offers significant advantages for various applications, including real-time recommendations, anomaly detection, and monitoring. The multi-way stream join operator facilitates the integration of multiple data streams into a single operator, allowing for a more comprehensive understanding by consolidating information from diverse sources. Although this operator is valuable in stream processing systems, its current probe order is determined prior to execution, making it challenging to adapt to real-time and unpredictable data streams, which can potentially diminish its operational efficiency. In this paper, we introduce a runtime-optimized multi-way stream join operator that incorporates various adaptive strategies to enhance the probe order during the joining of multi-way data streams. The operator's runtime operation is divided into cycles, during which relevant statistical information from the data streams is collected and updated. Historical statistical data is then utilized to predict the characteristics of the data streams in the current cycle using a quadratic exponential smoothing prediction method. An adaptive optimization algorithm based on a cost model, namely dpPick, is subsequently designed to refine the probe order, enabling better adaptation to real-time, unknown data streams and improving the operator's processing efficiency. Experiments conducted on the TPC-DS dataset demonstrate that the proposed multi-way stream join method significantly outperforms the comparative method in terms of processing efficiency.

cs.DB↗

Streaming SQL Multi-Way Join Method for Long State Streams

Streaming computing effectively manages large-scale streaming data in real-time, making it ideal for applications such as real-time recommendations, anomaly detection, and monitoring, all of which require immediate processing. In this context, the multi-way stream join operator is crucial, as it combines multiple data streams into a single operator, providing deeper insights through the integration of information from various sources. However, challenges related to memory limitations can arise when processing long state-based data streams, particularly in the area of streaming SQL. In this paper, we propose a streaming SQL multi-way stream join method that utilizes the LSM-Tree to address this issue. We first introduce a multi-way stream join operator called UMJoin, which employs an LSM-Tree state backend to leverage disk storage, thereby increasing the capacity for storing multi-way stream states beyond what memory can accommodate. Subsequently, we develop a method for converting execution plans, referred to as TSC, specifically for the UMJoin operator. This method identifies binary join tree patterns and generates corresponding multi-way stream join nodes, enabling us to transform execution plans based on binary joins into those that incorporate UMJoin nodes. This transformation facilitates the application of the UMJoin operator in streaming SQL. Experiments with the TPC-DS dataset demonstrate that the UMJoin operator can effectively process long state-based data streams, even with limited memory. Furthermore, tests on execution plan conversion for multi-way stream join queries using the TPC-H benchmark confirm the effectiveness of the TSC method in executing these conversions.

cs.DB↗

DCSim: Computing and Networking Integration based Container Scheduling Simulator for Data Centers

The increasing prevalence of cloud-native technologies, particularly containers, has led to the widespread adoption of containerized deployments in data centers. The advancement of deep neural network models has increased the demand for container-based distributed model training and inference, where frequent data transmission among nodes has emerged as a significant performance bottleneck. However, traditional container scheduling simulators often overlook the influence of network modeling on the efficiency of container scheduling, primarily concentrating on modeling computational resources. In this paper, we focus on a container scheduling simulator based on collaboration between computing and networking within data centers. We propose a new container scheduling simulator for data centers, named DCSim. The simulator consists of several modules: a data center module, a network simulation module, a container scheduling module, a discrete event-driven module, and a data collection and analysis module. Together, these modules provide heterogeneous computing power modeling and dynamic network simulation capabilities. We design a discrete event model using SimPy to represent various aspects of container processing, including container requests, scheduling, execution, pauses, communication, migration, and termination within data centers. Among these, lightweight virtualization technology based on Mininet is employed to construct a software-defined network. An experimental environment for container scheduling simulation was established, and functional and performance tests were conducted on the simulator to validate its scheduling simulation capabilities.

cs.DC↗

A Multicast Scheme for Live Streaming Courses in Large-Scale, Geographically Dense Campus Networks

Video courses have become a significant component of modern education. However, the increasing demand for live streaming video courses places considerable strain on the service capabilities of campus networks. The challenges associated with live streaming course videos in campus network environments exhibit distinct spatial distribution characteristics. The audience for specific video courses may be highly concentrated in certain areas, leading to a large number of users attempting to access the same live stream simultaneously. Utilizing a Content Delivery Network (CDN) to distribute videos in these campus scenarios creates substantial unicast pressure on edge CDN servers. This paper proposes a two-layer dynamic partitioning Recursive Bit String (RBS) virtual domain network layer multicast architecture specifically designed for large-scale, geographically dense multicast scenarios within campus networks. This approach reduces redundant multicast messages by approximately 10-30\% compared to the two-layer fixed partitioning method. Additionally, it establishes multicast source authentication capabilities based on Source Address Validation Improvement (SAVI) and facilitates secure multicast group key exchange using a concise exchange protocol within the WebRTC framework. In the next-generation data plane of programmable software-defined networks, the RBS stateless multicast technology can be integrated with the unique characteristics of large-scale, geographically dense campus network scenarios to dynamically and efficiently extend multicast coverage to every dormitory.

cs.NI↗

Transformer and Snowball Graph Convolution Learning for Brain functional network Classification

Advanced deep learning methods, especially graph neural networks (GNNs), are increasingly expected to learn from brain functional network data and predict brain disorders. In this paper, we proposed a novel Transformer and snowball encoding networks (TSEN) for brain functional network classification, which introduced Transformer architecture with graph snowball connection into GNNs for learning whole-graph representation. TSEN combined graph snowball connection with graph Transformer by snowball encoding layers, which enhanced the power to capture multi-scale information and global patterns of brain functional networks. TSEN also introduced snowball graph convolution as position embedding in Transformer structure, which was a simple yet effective method for capturing local patterns naturally. We evaluated the proposed model by two large-scale brain functional network datasets from autism spectrum disorder and major depressive disorder respectively, and the results demonstrated that TSEN outperformed the state-of-the-art GNN models and the graph-transformer based GNN models.

cs.LG↗

BrainNPT: Pre-training of Transformer networks for brain network classification

Deep learning methods have advanced quickly in brain imaging analysis over the past few years, but they are usually restricted by the limited labeled data. Pre-trained model on unlabeled data has presented promising improvement in feature learning in many domains, including natural language processing and computer vision. However, this technique is under-explored in brain network analysis. In this paper, we focused on pre-training methods with Transformer networks to leverage existing unlabeled data for brain functional network classification. First, we proposed a Transformer-based neural network, named as BrainNPT, for brain functional network classification. The proposed method leveraged token as a classification embedding vector for the Transformer model to effectively capture the representation of brain network. Second, we proposed a pre-training framework for BrainNPT model to leverage unlabeled brain network data to learn the structure information of brain networks. The results of classification experiments demonstrated the BrainNPT model without pre-training achieved the best performance with the state-of-the-art models, and the BrainNPT model with pre-training strongly outperformed the state-of-the-art models. The pre-training BrainNPT model improved 8.75% of accuracy compared with the model without pre-training. We further compared the pre-training strategies, analyzed the influence of the parameters of the model, and interpreted the trained model.

q-bio.NC↗

Secure two-way fiber-optic time transfer against sub-ns asymmetric delay attack

Two-way fiber-optic time transfer is a promising precise time synchronization technique with sub-nanosecond accuracy. However, asymmetric delay attack is a serious threat which cannot be prevent by any encryption method. In this paper, a dynamic model based scheme is proposed to defense the sub-nanosecond asymmetric delay attack. A threshold is set according to the estimated time difference by a two-state clock model where the fixed frequency difference is excluded from the time difference to detect the asymmetric delay attack which is smaller than the time difference induced by the fixed frequency difference. Theoretical simulation and experimental demonstration are implemented to prove the feasibility of the scheme. A two-way fiber-optic time transfer system with time stability with 24.5ps, 3.98ps, and 2.95ps at 1s, 10s, and 100s averaging time is shown under sub-ns asymmetric time delay attack experimentally. The proposed method provides a promising secure sub-ns precise time synchronization technique against asymmetric delay attack.

quant-ph↗

A Multi-modal Fusion Framework Based on Multi-task Correlation Learning for Cancer Prognosis Prediction

Morphological attributes from histopathological images and molecular profiles from genomic data are important information to drive diagnosis, prognosis, and therapy of cancers. By integrating these heterogeneous but complementary data, many multi-modal methods are proposed to study the complex mechanisms of cancers, and most of them achieve comparable or better results from previous single-modal methods. However, these multi-modal methods are restricted to a single task (e.g., survival analysis or grade classification), and thus neglect the correlation between different tasks. In this study, we present a multi-modal fusion framework based on multi-task correlation learning (MultiCoFusion) for survival analysis and cancer grade classification, which combines the power of multiple modalities and multiple tasks. Specifically, a pre-trained ResNet-152 and a sparse graph convolutional network (SGCN) are used to learn the representations of histopathological images and mRNA expression data respectively. Then these representations are fused by a fully connected neural network (FCNN), which is also a multi-task shared network. Finally, the results of survival analysis and cancer grade classification output simultaneously. The framework is trained by an alternate scheme. We systematically evaluate our framework using glioma datasets from The Cancer Genome Atlas (TCGA). Results demonstrate that MultiCoFusion learns better representations than traditional feature extraction methods. With the help of multi-task alternating learning, even simple multi-modal concatenation can achieve better performance than other deep learning and traditional methods. Multi-task learning can improve the performance of multiple tasks not just one of them, and it is effective in both single-modal and multi-modal data.

cs.LG↗

AEFE: Automatic Embedded Feature Engineering for Categorical Features

The challenge of solving data mining problems in e-commerce applications such as recommendation system (RS) and click-through rate (CTR) prediction is how to make inferences by constructing combinatorial features from a large number of categorical features while preserving the interpretability of the method. In this paper, we propose Automatic Embedded Feature Engineering(AEFE), an automatic feature engineering framework for representing categorical features, which consists of various components including custom paradigm feature construction and multiple feature selection. By selecting the potential field pairs intelligently and generating a series of interpretable combinatorial features, our framework can provide a set of unseen generated features for enhancing model performance and then assist data analysts in discovering the feature importance for particular data mining tasks. Furthermore, AEFE is distributed implemented by task-parallelism, data sampling, and searching schema based on Matrix Factorization field combination, to optimize the performance and enhance the efficiency and scalability of the framework. Experiments conducted on some typical e-commerce datasets indicate that our method outperforms the classical machine learning models and state-of-the-art deep learning models.

cs.LG↗

MESD: Exploring Optical Flow Assessment on Edge of Motion Objects with Motion Edge Structure Difference

The optical flow estimation has been assessed in various applications. In this paper, we propose a novel method named motion edge structure difference(MESD) to assess estimation errors of optical flow fields on edge of motion objects. We implement comparison experiments for MESD by evaluating five representative optical flow algorithms on four popular benchmarks: MPI Sintel, Middlebury, KITTI 2012 and KITTI 2015. Our experimental results demonstrate that MESD can reasonably and discriminatively assess estimation errors of optical flow fields on motion edge. The results indicate that MESD could be a supplementary metric to existing general assessment metrics for evaluating optical flow algorithms in related computer vision applications.

cs.CV↗

An end-to-end Generative Retrieval Method for Sponsored Search Engine --Decoding Efficiently into a Closed Target Domain

In this paper, we present a generative retrieval method for sponsored search engine, which uses neural machine translation (NMT) to generate keywords directly from query. This method is completely end-to-end, which skips query rewriting and relevance judging phases in traditional retrieval systems. Different from standard machine translation, the target space in the retrieval setting is a constrained closed set, where only committed keywords should be generated. We present a Trie-based pruning technique in beam search to address this problem. The biggest challenge in deploying this method into a real industrial environment is the latency impact of running the decoder. Self-normalized training coupled with Trie-based dynamic pruning dramatically reduces the inference time, yielding a speedup of more than 20 times. We also devise an mixed online-offline serving architecture to reduce the latency and CPU consumption. To encourage the NMT to generate new keywords uncovered by the existing system, training data is carefully selected. This model has been successfully applied in Baidu's commercial search engine as a supplementary retrieval branch, which has brought a remarkable revenue improvement of more than 10 percents.

cs.IR↗