arXiv subjects

Wentong Cai

Publications and source records attributed to Wentong Cai.

16 recordsLinked to original sources

arXiv2026

Efficient Learned Data Compression via Dual-Stream Feature Decoupling

While Learned Data Compression (LDC) has achieved superior compression ratios, balancing precise probability modeling with system efficiency remains challenging. Crucially, uniform single-stream architectures struggle to simultaneously capture micro-syntactic and macro-semantic features, necessitating deep serial stacking that exacerbates latency. Compounding this, heterogeneous systems are constrained by device speed mismatches, where throughput is capped by Amdahl's Law due to serial processing. To this end, we propose a Dual-Stream Multi-Scale Decoupler that disentangles local and global contexts to replace deep serial processing with shallow parallel streams, and incorporate a Hierarchical Gated Refiner for adaptive feature refinement and precise probability modeling. Furthermore, we design a Concurrent Stream-Parallel Pipeline, which overcomes systemic bottlenecks to achieve full-pipeline parallelism. Extensive experiments demonstrate that our method achieves state-of-the-art performance in both compression ratio and throughput, while maintaining the lowest latency and memory usage. The code is available at https://github.com/huidong-ma/FADE.

cs.CL↗

arXiv2026

Dimensional Peeking for Low-Variance Gradients in Zeroth-Order Discrete Optimization via Simulation

Gradient-based optimization methods are commonly used to identify local optima in high-dimensional spaces. When derivatives cannot be evaluated directly, stochastic estimators can provide approximate gradients. However, these estimators' perturbation-based sampling of the objective function introduces variance that can lead to slow convergence. In this paper, we present dimensional peeking, a variance reduction method for gradient estimation in discrete optimization via simulation. By lifting the sampling granularity from scalar values to classes of values that follow the same control flow path, we increase the information gathered per simulation evaluation. Our derivation from an established smoothed gradient estimator shows that the method does not introduce any bias. We present an implementation via a custom numerical data type to transparently carry out dimensional peeking over C++ programs. Variance reductions by factors of up to 7.9 are observed for three simulation-based optimization problems with high-dimensional input. The optimization progress compared to three meta-heuristics shows that dimensional peeking increases the competitiveness of zeroth-order optimization for discrete and non-convex simulations.

cs.LG↗

arXiv2026

AgentGC: Evolutionary Learning-based Lossless Compression for Genomics Data with LLM-driven Multiple Agent

Lossless compression has made significant advancements in Genomics Data (GD) storage, sharing and management. Current learning-based methods are non-evolvable with problems of low-level compression modeling, limited adaptability, and user-unfriendly interface. To this end, we propose AgentGC, the first evolutionary Agent-based GD Compressor, consisting of 3 layers with multi-agent named Leader and Worker. Specifically, the 1) User layer provides a user-friendly interface via Leader combined with LLM; 2) Cognitive layer, driven by the Leader, integrates LLM to consider joint optimization of algorithm-dataset-system, addressing the issues of low-level modeling and limited adaptability; and 3) Compression layer, headed by Worker, performs compression & decompression via a automated multi-knowledge learning-based compression framework. On top of AgentGC, we design 3 modes to support diverse scenarios: CP for compression-ratio priority, TP for throughput priority, and BM for balanced mode. Compared with 14 baselines on 9 datasets, the average compression ratios gains are 16.66%, 16.11%, and 16.33%, the throughput gains are 4.73x, 9.23x, and 9.15x, respectively.

cs.AI↗

arXiv2025

CompARE: A Computational framework for Airborne Respiratory disease Evaluation integrating flow physics and human behavior

The risk of indoor airborne transmission among co-located individuals is generally non-uniform, which remains a critical challenge for public health modelling. Thus, we present CompARE, an integrated risk assessment framework for indoor airborne disease transmission that reveals a striking bimodal distribution of infection risk driven by airflow dynamics and human behavior. Combining computational fluid dynamics (CFD), machine learning (ML), and agent-based modeling (ABM), our model captures the complex interplay between aerosol transport, human mobility, and environmental context. Based on a prototypical childcare center, our approach quantifies how incorporation of ABM can unveil significantly different infection risk profiles across agents, with more than two-fold change in risk of infection between the individuals with the lowest and highest risks in more than 90% of cases, despite all individuals being in the same overall environment. We found that infection risk distributions can exhibit not only a striking bimodal pattern in certain activities but also exponential decay and fat-tailed behavior in others. Specifically, we identify low-risk modes arising from source containment, as well as high-risk tails from prolonged close contact. Our approach enables near-real-time scenario analysis and provides policy-relevant quantitative insights into how ventilation design, spatial layout, and social distancing policies can mitigate transmission risk. These findings challenge simple distance-based heuristics and support the design of targeted, evidence-based interventions in high-occupancy indoor settings.

physics.soc-ph↗

arXiv2025

Learning to Optimize Capacity Planning in Semiconductor Manufacturing

In manufacturing, capacity planning is the process of allocating production resources in accordance with variable demand. The current industry practice in semiconductor manufacturing typically applies heuristic rules to prioritize actions, such as future change lists that account for incoming machine and recipe dedications. However, while offering interpretability, heuristics cannot easily account for the complex interactions along the process flow that can gradually lead to the formation of bottlenecks. Here, we present a neural network-based model for capacity planning on the level of individual machines, trained using deep reinforcement learning. By representing the policy using a heterogeneous graph neural network, the model directly captures the diverse relationships among machines and processing steps, allowing for proactive decision-making. We describe several measures taken to achieve sufficient scalability to tackle the vast space of possible machine-level actions. Our evaluation results cover Intel's small-scale Minifab model and preliminary experiments using the popular SMT2020 testbed. In the largest tested scenario, our trained policy increases throughput and decreases cycle time by about 1.8% each.

cs.LG↗

arXiv2025

PMKLC: Parallel Multi-Knowledge Learning-based Lossless Compression for Large-Scale Genomics Database

Learning-based lossless compressors play a crucial role in large-scale genomic database backup, storage, transmission, and management. However, their 1) inadequate compression ratio, 2) low compression \& decompression throughput, and 3) poor compression robustness limit their widespread adoption and application in both industry and academia. To solve those challenges, we propose a novel \underline{P}arallel \underline{M}ulti-\underline{K}nowledge \underline{L}earning-based \underline{C}ompressor (PMKLC) with four crucial designs: 1) We propose an automated multi-knowledge learning-based compression framework as compressors' backbone to enhance compression ratio and robustness; 2) we design a GPU-accelerated ($s$,$k$)-mer encoder to optimize compression throughput and computing resource usage; 3) we introduce data block partitioning and Step-wise Model Passing (SMP) mechanisms for parallel acceleration; 4) We design two compression modes PMKLC-S and PMKLC-M to meet the complex application scenarios, where the former runs on a resource-constrained single GPU and the latter is multi-GPU accelerated. We benchmark PMKLC-S/M and 14 baselines (7 traditional and 7 leaning-based) on 15 real-world datasets with different species and data sizes. Compared to baselines on the testing datasets, PMKLC-S/M achieve the average compression ratio improvement up to 73.609\% and 73.480\%, the average throughput improvement up to 3.036$\times$ and 10.710$\times$, respectively. Besides, PMKLC-S/M also achieve the best robustness and competitive memory cost, indicating its greater stability against datasets with different probability distribution perturbations, and its strong ability to run on memory-constrained devices.

cs.LG↗

arXiv2025

GenAI for Simulation Model in Model-Based Systems Engineering

Generative AI (GenAI) has demonstrated remarkable capabilities in code generation, and its integration into complex product modeling and simulation code generation can significantly enhance the efficiency of the system design phase in Model-Based Systems Engineering (MBSE). In this study, we introduce a generative system design methodology framework for MBSE, offering a practical approach for the intelligent generation of simulation models for system physical properties. First, we employ inference techniques, generative models, and integrated modeling and simulation languages to construct simulation models for system physical properties based on product design documents. Subsequently, we fine-tune the language model used for simulation model generation on an existing library of simulation models and additional datasets generated through generative modeling. Finally, we introduce evaluation metrics for the generated simulation models for system physical properties. Our proposed approach to simulation model generation presents the innovative concept of scalable templates for simulation models. Using these templates, GenAI generates simulation models for system physical properties through code completion. The experimental results demonstrate that, for mainstream open-source Transformer-based models, the quality of the simulation model is significantly improved using the simulation model generation method proposed in this paper.

cs.SE↗

arXiv2024

Characterizing pedestrian contact interaction trajectories to understand spreading risk in human crowds

A spreading process can be observed when particular information, substances, or diseases spread through a population over time in social and biological systems. It is widely believed that contact interactions among individual entities play an essential role in the spreading process. Although contact interactions are often influenced by geometrical conditions, little attention has been paid to understand their effects, especially on contact duration among pedestrians. To examine how the pedestrian flow setups affect contact duration distribution, we have analyzed trajectories of pedestrians in contact interactions collected from pedestrian flow experiments of uni-, bi- and multi-directional setups. Based on turning angle entropy and efficiency, we have classified the type of motion observed in the contact interactions. We have found that the majority of contact interactions in the unidirectional flow setup can be categorized as confined motion, hinting at the possibility of long-lived contact duration. However, ballistic motion is more frequently observed in the other flow conditions, yielding frequent, brief contact interactions. Our results demonstrate that observing more confined motions is likely associated with the increase of parallel contact interactions regardless of pedestrian flow setups. This study highlights that the confined motions tend to yield longer contact duration, suggesting that the infectious disease transmission risk would be considerable even for low transmissibility. These results have important implications for crowd management in the context of minimizing spreading risk. This work is an extended version of Kwak et al. (2023) presented at the 2023 International Conference on Computational Science (ICCS).

physics.soc-ph↗

arXiv2024

Characterization of pedestrian contact interaction trajectories

A spreading process can be observed when a particular behavior, substance, or disease spreads through a population over time in social and biological systems. It is widely believed that contact interactions among individual entities play an essential role in the spreading process. Although the contact interactions are often influenced by geometrical conditions, little attention has been paid to understand their effects especially on contact duration among pedestrians. To examine how the pedestrian flow setups affect contact duration distribution, we have analyzed trajectories of pedestrians in contact interactions collected from pedestrian flow experiments of uni-, bi- and multi-directional setups. Based on standardized maximal distance, we have classified types of motions observed in the contact interactions. We have found that almost all motion in the unidirectional flow setup can be characterized as subdiffusive motion, suggesting that the empirically measured contact duration tends to be longer than one estimated by ballistic motion assumption. However, Brownian motion is more frequently observed from other flow setups, indicating that the contact duration estimated by ballistic motion assumption shows good agreement with the empirically measured one. Furthermore, when the difference in relative speed distributions between the experimental data and ballistic motion assumption is larger, more subdiffusive motions are observed. This study also has practical implications. For instance, it highlights that geometrical conditions yielding smaller difference in the relative speed distributions are preferred when diseases can be transmitted through face-to-face interactions.

physics.soc-ph↗

arXiv2022

Crowd-level Abnormal Behavior Detection via Multi-scale Motion Consistency Learning

Detecting abnormal crowd motion emerging from complex interactions of individuals is paramount to ensure the safety of crowds. Crowd-level abnormal behaviors (CABs), e.g., counter flow and crowd turbulence, are proven to be the crucial causes of many crowd disasters. In the recent decade, video anomaly detection (VAD) techniques have achieved remarkable success in detecting individual-level abnormal behaviors (e.g., sudden running, fighting and stealing), but research on VAD for CABs is rather limited. Unlike individual-level anomaly, CABs usually do not exhibit salient difference from the normal behaviors when observed locally, and the scale of CABs could vary from one scenario to another. In this paper, we present a systematic study to tackle the important problem of VAD for CABs with a novel crowd motion learning framework, multi-scale motion consistency network (MSMC-Net). MSMC-Net first captures the spatial and temporal crowd motion consistency information in a graph representation. Then, it simultaneously trains multiple feature graphs constructed at different scales to capture rich crowd patterns. An attention network is used to adaptively fuse the multi-scale features for better CAB detection. For the empirical study, we consider three large-scale crowd event datasets, UMN, Hajj and Love Parade. Experimental results show that MSMC-Net could substantially improve the state-of-the-art performance on all the datasets.

cs.CV↗

arXiv2021

Influence of committed volunteers on helping behavior in emergency evacuations

We study how the presence of committed volunteers influences the collective helping behavior in emergency evacuation scenarios. In this study, committed volunteers do not change their decision to help injured persons, implying that other evacuees may adapt their helping behavior through strategic interactions. An evolutionary game theoretic model is developed which is then coupled to a pedestrian movement model to examine the collective helping behavior in evacuations. By systematically controlling the number of committed volunteers and payoff parameters, we have characterized and summarized various collective helping behaviors in phase diagrams. From our numerical simulations, we observe that the existence of committed volunteers can promote cooperation but adding additional committed volunteers is effective only above a minimum number of committed volunteers. This study also highlights that the evolution of collective helping behavior is strongly affected by the evacuation process.

physics.soc-ph↗

arXiv2021

Time-Efficient and High-Quality Graph Partitioning for Graph Dynamic Scaling

The dynamic scaling of distributed computations plays an important role in the utilization of elastic computational resources, such as the cloud. It enables the provisioning and de-provisioning of resources to match dynamic resource availability and demands. In the case of distributed graph processing, changing the number of the graph partitions while maintaining high partitioning quality imposes serious computational overheads as typically a time-consuming graph partitioning algorithm needs to execute each time repartitioning is required. In this paper, we propose a dynamic scaling method that can efficiently change the number of graph partitions while keeping its quality high. Our idea is based on two techniques: preprocessing and very fast edge partitioning, called graph edge ordering and chunk-based edge partitioning, respectively. The former converts the graph data into an ordered edge list in such a way that edges with high locality are closer to each other. The latter immediately divides the ordered edge list into an arbitrary number of high-quality partitions. The evaluation with the real-world billion-scale graphs demonstrates that our proposed approach significantly reduces the repartitioning time, while the partitioning quality it achieves is on par with that of the best existing static method.

cs.DC↗

arXiv2020

Estimating horizontal movement performance of patient beds and the impact on emergency evacuation time

Emergency evacuation of patients from a hospital can be challenging in the event of a fire. Most emergency evacuation studies are based on the assumption that pedestrians are ambulant and can egress by themselves. However, this is often not the case during emergency evacuations in healthcare facilities such as hospitals and nursing homes. To investigate emergency evacuations in such healthcare facilities, we performed a series of controlled experiments to study the dynamics of patient beds in horizontal movement. We considered a patient bed because it is one of the commonly used devices to transport patients within healthcare facilities. Through a series of controlled experiments, we examined the change of velocity in corner turning movements and speed reductions in multiple trips between both ends of a straight corridor. Based on the experimental results, we then developed a mathematical model of total evacuation time prediction for a patient bed horizontally moving in a healthcare facility. Factoring uncertainty in the horizontal movement, we produced the probability distribution of movement duration and estimated the probability that an evacuation can be safely performed within certain amount of time. In addition, we predicted that the evacuation time would be longer than the prediction results from an existing model which assumes constant movement speed. Our results from the model demonstrated good agreement with our experimental results.

physics.soc-ph↗

arXiv2020

Modeling Helping Behavior in Emergency Evacuations Using Volunteer's Dilemma Game

People often help others who are in trouble, especially in emergency evacuation situations. For instance, during the 2005 London bombings, it was reported that evacuees helped injured persons to escape the place of danger. In terms of game theory, it can be understood that such helping behavior provides a collective good while it is a costly behavior because the volunteers spend extra time to assist the injured persons in case of emergency evacuations. In order to study the collective effects of helping behavior in emergency evacuations, we have performed numerical simulations of helping behavior among evacuees in a room evacuation scenario. Our simulation model is based on the volunteer's dilemma game reflecting volunteering cost. The game theoretic model is coupled with a social force model to understand the relationship between the spatial and social dynamics of evacuation scenarios. By systematically changing the cost parameter of helping behavior, we observed different patterns of collective helping behaviors and these collective patterns are summarized with a phase diagram.

physics.soc-ph↗

arXiv2019

Distributed Edge Partitioning for Trillion-edge Graphs

We propose Distributed Neighbor Expansion (Distributed NE), a parallel and distributed graph partitioning method that can scale to trillion-edge graphs while providing high partitioning quality. Distributed NE is based on a new heuristic, called parallel expansion, where each partition is constructed in parallel by greedily expanding its edge set from a single vertex in such a way that the increase of the vertex cuts becomes local minimal. We theoretically prove that the proposed method has the upper bound in the partitioning quality. The empirical evaluation with various graphs shows that the proposed method produces higher-quality partitions than the state-of-the-art distributed graph partitioning algorithms. The performance evaluation shows that the space efficiency of the proposed method is an order-of-magnitude better than the existing algorithms, keeping its time efficiency comparable. As a result, Distributed NE can partition a trillion-edge graph using only 256 machines within 70 minutes.

cs.DC↗

arXiv2018

A Survey on Agent-based Simulation using Hardware Accelerators

Due to decelerating gains in single-core CPU performance, computationally expensive simulations are increasingly executed on highly parallel hardware platforms. Agent-based simulations, where simulated entities act with a certain degree of autonomy, frequently provide ample opportunities for parallelisation. Thus, a vast variety of approaches proposed in the literature demonstrated considerable performance gains using hardware platforms such as many-core CPUs and GPUs, merged CPU-GPU chips as well as FPGAs. Typically, a combination of techniques is required to achieve high performance for a given simulation model, putting substantial burden on modellers. To the best of our knowledge, no systematic overview of techniques for agent-based simulations on hardware accelerators has been given in the literature. To close this gap, we provide an overview and categorisation of the literature according to the applied techniques. Since at the current state of research, challenges such as the partitioning of a model for execution on heterogeneous hardware are still a largely manual process, we sketch directions for future research towards automating the hardware mapping and execution. This survey targets modellers seeking an overview of suitable hardware platforms and execution techniques for a specific simulation model, as well as methodology researchers interested in potential research gaps requiring further exploration.

cs.MA↗