Publications (45)
Towards Robust Training of Neural Networks by Regularizing Adversarial Gradients
Fuxun Yu, Zirui Xu, Yanzhi Wang +2
In recent years, neural networks have demonstrated outstanding effectiveness in a large amount of applications.However, recent works have shown that neural networks are susceptible…
ADMM for Efficient Deep Learning with Global Convergence
Junxiang Wang, Fuxun Yu, Xiang Chen +1
Alternating Direction Method of Multipliers (ADMM) has been used successfully in many conventional machine learning applications and is considered to be a useful alternative to Sto…
VULPO: Context-Aware Vulnerability Detection via On-Policy LLM Optimization
Youpeng Li, Fuxun Yu, Weiliang Qi +1
Large language models (LLMs) have recently shown strong potential in vulnerability detection (VD). However, accurately detecting vulnerabilities in real-world repositories requires…
Functionality-Oriented Convolutional Filter Pruning
Zhuwei Qin, Fuxun Yu, Chenchen Liu +1
The sophisticated structure of Convolutional Neural Network (CNN) allows for outstanding performance, but at the cost of intensive computation. As significant redundancies inevitab…
Revisiting Pre-trained Language Models for Vulnerability Detection
Youpeng Li, Weiliang Qi, Xuyu Wang +2
The rapid advancement of pre-trained language models (PLMs) has demonstrated promising results for various code-related tasks. However, their effectiveness in detecting real-world…
Towards Latency-aware DNN Optimization with GPU Runtime Analysis and Tail Effect Elimination
Fuxun Yu, Zirui Xu, Tong Shen +12
Despite the superb performance of State-Of-The-Art (SOTA) DNNs, the increasing computational cost makes them very challenging to meet real-time latency and accuracy requirements. A…
Interpreting and Evaluating Neural Network Robustness
Fuxun Yu, Zhuwei Qin, Chenchen Liu +3
Recently, adversarial deception becomes one of the most considerable threats to deep neural networks. However, compared to extensive research in new designs of various adversarial…
Progressive Weight Pruning of Deep Neural Networks using ADMM
Shaokai Ye, Tianyun Zhang, Kaiqi Zhang +10
Deep neural networks (DNNs) although achieving human-level performance in many domains, have very large model size that hinders their broader applications on edge computing devices…
DoPa: A Comprehensive CNN Detection Methodology against Physical Adversarial Attacks
Zirui Xu, Fuxun Yu, Xiang Chen
Recently, Convolutional Neural Networks (CNNs) demonstrate a considerable vulnerability to adversarial attacks, which can be easily misled by adversarial perturbations. With more a…
Online Learning via Memory: Retrieval-Augmented Detector Adaptation
Yanan Jian, Fuxun Yu, Qi Zhang +3
This paper presents a novel way of online adapting any off-the-shelf object detection model to a novel domain without retraining the detector model. Inspired by how humans quickly…
Efficient Neural Network Implementation with Quadratic Neuron
Zirui Xu, Jinjun Xiong, Fuxun Yu +1
Previous works proved that the combination of the linear neuron network with nonlinear activation functions (e.g. ReLu) can achieve nonlinear function approximation. However, simpl…
How convolutional neural network see the world - A survey of convolutional neural network visualization methods
Zhuwei Qin, Fuxun Yu, Chenchen Liu +1
Nowadays, the Convolutional Neural Networks (CNNs) have achieved impressive performance on many computer vision related tasks, such as object detection, image recognition, image re…
From SFT to RL: Demystifying the Post-Training Pipeline for LLM-based Vulnerability Detection
Youpeng Li, Fuxun Yu, Xinda Wang
The integration of LLMs into vulnerability detection (VD) has shifted the field toward more interpretable and context-aware analysis. While post-training techniques have shown prom…
QuadraNet: Improving High-Order Neural Interaction Efficiency with Hardware-Aware Quadratic Neural Networks
Chenhui Xu, Fuxun Yu, Zirui Xu +3
Recent progress in computer vision-oriented neural network designs is mostly driven by capturing high-order neural interactions among inputs and features. And there emerged a varie…
Out-of-Distribution Detection via Deep Multi-Comprehension Ensemble
Chenhui Xu, Fuxun Yu, Zirui Xu +2
Recent research underscores the pivotal role of the Out-of-Distribution (OOD) feature representation field scale in determining the efficacy of models in OOD detection. Consequentl…
Multi-Agent Geospatial Copilots for Remote Sensing Workflows
Chaehong Lee, Varatheepan Paramanayakam, Andreas Karatzas +7
We present GeoLLM-Squad, a geospatial Copilot that introduces the novel multi-agent paradigm to remote sensing (RS) workflows. Unlike existing single-agent approaches that rely on…
Tiny but Accurate: A Pruned, Quantized and Optimized Memristor Crossbar Framework for Ultra Efficient DNN Implementation
Xiaolong Ma, Geng Yuan, Sheng Lin +6
The state-of-art DNN structures involve intensive computation and high memory storage. To mitigate the challenges, the memristor crossbar array has emerged as an intrinsically suit…
Fed2: Feature-Aligned Federated Learning
Fuxun Yu, Weishan Zhang, Zhuwei Qin +5
Federated learning learns from scattered data by fusing collaborative models from local nodes. However, the conventional coordinate-based model averaging by FedAvg ignored the rand…
HASP: A High-Performance Adaptive Mobile Security Enhancement Against Malicious Speech Recognition
Zirui Xu, Fuxun Yu, Chenchen Liu +1
Nowadays, machine learning based Automatic Speech Recognition (ASR) technique has widely spread in smartphones, home devices, and public facilities. As convenient as this technolog…
LanCe: A Comprehensive and Lightweight CNN Defense Methodology against Physical Adversarial Attacks on Embedded Multimedia Applications
Zirui Xu, Fuxun Yu, Xiang Chen
Recently, adversarial attacks can be applied to the physical world, causing practical issues to various Convolutional Neural Networks (CNNs) powered applications. Most existing phy…
Unlocking Zero-Shot Geospatial Reasoning via Indirect Rewards
Chenhui Xu, Fuxun Yu, Michael J. Bianco +15
Training robust reasoning vision-language models (VLMs) in rare domains (such as geospatial) is fundamentally constrained by supervision scarcity. While raw geospatial imagery is a…
Stable Diffusion For Aerial Object Detection
Yanan Jian, Fuxun Yu, Simranjit Singh +1
Aerial object detection is a challenging task, in which one major obstacle lies in the limitations of large-scale data collection and the long-tail distribution of certain classes.…
QuadraLib: A Performant Quadratic Neural Network Library for Architecture Optimization and Design Exploration
Zirui Xu, Fuxun Yu, Jinjun Xiong +1
The significant success of Deep Neural Networks (DNNs) is highly promoted by the multiple sophisticated DNN libraries. On the contrary, although some work have proved that Quadrati…
Unsupervised Domain Adaptation for Object Detection via Cross-Domain Semi-Supervised Learning
Fuxun Yu, Di Wang, Yinpeng Chen +7
Current state-of-the-art object detectors can have significant performance drop when deployed in the wild due to domain gaps with training data. Unsupervised Domain Adaptation (UDA…
Third ArchEdge Workshop: Exploring the Design Space of Efficient Deep Neural Networks
Fuxun Yu, Dimitrios Stamoulis, Di Wang +2
This paper gives an overview of our ongoing work on the design space exploration of efficient deep neural networks (DNNs). Specifically, we cover two aspects: (1) static architectu…
ASP:A Fast Adversarial Attack Example Generation Framework based on Adversarial Saliency Prediction
Fuxun Yu, Qide Dong, Xiang Chen
With the excellent accuracy and feasibility, the Neural Networks have been widely applied into the novel intelligent applications and systems. However, with the appearance of the A…
QuadraNet V2: Efficient and Sustainable Training of High-Order Neural Networks with Quadratic Adaptation
Chenhui Xu, Xinyao Wang, Fuxun Yu +2
Machine learning is evolving towards high-order models that necessitate pre-training on extensive datasets, a process associated with significant overheads. Traditional models, des…
GACER: Granularity-Aware ConcurrEncy Regulation for Multi-Tenant Deep Learning
Yongbo Yu, Fuxun Yu, Mingjia Zhang +4
As deep learning continues to advance and is applied to increasingly complex scenarios, the demand for concurrent deployment of multiple neural network models has arisen. This dema…
Task-Adaptive Incremental Learning for Intelligent Edge Devices
Zhuwei Qin, Fuxun Yu, Xiang Chen
Convolutional Neural Networks (CNNs) are used for a wide range of image-related tasks such as image classification and object detection. However, a large pre-trained CNN model cont…
Demystifying Neural Network Filter Pruning
Zhuwei Qin, Fuxun Yu, ChenChen Liu +1
Based on filter magnitude ranking (e.g. L1 norm), conventional filter pruning methods for Convolutional Neural Networks (CNNs) have been proved with great effectiveness in computat…
A Survey of Large-Scale Deep Learning Serving System Optimization: Challenges and Opportunities
Fuxun Yu, Di Wang, Longfei Shangguan +4
Deep Learning (DL) models have achieved superior performance in many application domains, including vision, language, medical, commercial ads, entertainment, etc. With the fast dev…
Helios: Heterogeneity-Aware Federated Learning with Dynamically Balanced Collaboration
Zirui Xu, Fuxun Yu, Jinjun Xiong +1
In this paper, we propose Helios, a heterogeneity-aware FL framework to tackle the straggler issue. Helios identifies individual devices' heterogeneous training capability, and the…
Geospatial Foundational Embedder: Top-1 Winning Solution on EarthVision Embed2Scale Challenge (CVPR 2025)
Zirui Xu, Raphael Tang, Mike Bianco +4
EarthVision Embed2Scale challenge (CVPR 2025) aims to develop foundational geospatial models to embed SSL4EO-S12 hyperspectral geospatial data cubes into embedding vectors that fac…
Rollback-Free Stable Brick Structures Generation
Chenhui Xu, Ziyue Bai, Fuxun Yu +2
While autoregressive models have advanced 3D generation, creating physically stable brick structures remains a challenge due to the strict requirements of gravity and interconnecti…
Distilling Critical Paths in Convolutional Neural Networks
Fuxun Yu, Zhuwei Qin, Xiang Chen
Neural network compression and acceleration are widely demanded currently due to the resource constraints on most deployment targets. In this paper, through analyzing the filter ac…
Supporting Massive DLRM Inference Through Software Defined Memory
Ehsan K. Ardestani, Changkyu Kim, Seung Jae Lee +17
Deep Learning Recommendation Models (DLRM) are widespread, account for a considerable data center footprint, and grow by more than 1.5x per year. With model size soon to be in tera…
FedCAP: Robust Federated Learning via Customized Aggregation and Personalization
Youpeng Li, Xinda Wang, Fuxun Yu +3
Federated learning (FL), an emerging distributed machine learning paradigm, has been applied to various privacy-preserving scenarios. However, due to its distributed nature, FL fac…
Infinite-Dimensional Feature Interaction
Chenhui Xu, Fuxun Yu, Maoliang Li +4
The past neural network design has largely focused on feature representation space dimension and its capacity scaling (e.g., width, depth), but overlooked the feature interaction s…
A Survey of Multi-Tenant Deep Learning Inference on GPU
Fuxun Yu, Di Wang, Longfei Shangguan +3
Deep Learning (DL) models have achieved superior performance. Meanwhile, computing hardware like NVIDIA GPUs also demonstrated strong computing scaling trends with 2x throughput an…
Heterogeneous Federated Learning
Fuxun Yu, Weishan Zhang, Zhuwei Qin +5
Federated learning learns from scattered data by fusing collaborative models from local nodes. However, due to chaotic information distribution, the model fusion may suffer from st…
FedHC: A Scalable Federated Learning Framework for Heterogeneous and Resource-Constrained Clients
Min Zhang, Fuxun Yu, Yongbo Yu +3
Federated Learning (FL) is a distributed learning paradigm that empowers edge devices to collaboratively learn a global model leveraging local data. Simulating FL on GPU is essenti…
Automated Runtime-Aware Scheduling for Multi-Tenant DNN Inference on GPU
Fuxun Yu, Shawn Bray, Di Wang +4
With the fast development of deep neural networks (DNNs), many real-world applications are adopting multiple models to conduct compound tasks, such as co-running classification, de…
LLM-dCache: Improving Tool-Augmented LLMs with GPT-Driven Localized Data Caching
Simranjit Singh, Michael Fore, Andreas Karatzas +6
As Large Language Models (LLMs) broaden their capabilities to manage thousands of API calls, they are confronted with complex data operations across vast datasets with significant…
Interpreting Adversarial Robustness: A View from Decision Surface in Input Space
Fuxun Yu, Chenchen Liu, Yanzhi Wang +2
One popular hypothesis of neural network generalization is that the flat local minima of loss surface in parameter space leads to good generalization. However, we demonstrate that…
AntiDote: Attention-based Dynamic Optimization for Neural Network Runtime Efficiency
Fuxun Yu, Chenchen Liu, Di Wang +2
Convolutional Neural Networks (CNNs) achieved great cognitive performance at the expense of considerable computation load. To relieve the computation load, many optimization works…