Xu Chen
YOU?
Author Swipe
View article: Experimental and analytical investigation on continuous two-span RC beams flexural strengthened with aluminum alloy bars
Experimental and analytical investigation on continuous two-span RC beams flexural strengthened with aluminum alloy bars Open
View article: CoEdge-RAG: Optimizing Hierarchical Scheduling for Retrieval-Augmented LLMs in Collaborative Edge Computing
CoEdge-RAG: Optimizing Hierarchical Scheduling for Retrieval-Augmented LLMs in Collaborative Edge Computing Open
Motivated by the imperative for real-time responsiveness and data privacy preservation, large language models (LLMs) are increasingly deployed on resource-constrained edge devices to enable localized inference. To improve output quality, r…
View article: Poster: A Unified Framework for Simultaneous Video Analytics and Streaming on UAVs
Poster: A Unified Framework for Simultaneous Video Analytics and Streaming on UAVs Open
View article: STADI: Fine-Grained Step-Patch Diffusion Parallelism for Heterogeneous GPUs
STADI: Fine-Grained Step-Patch Diffusion Parallelism for Heterogeneous GPUs Open
The escalating adoption of diffusion models for applications such as image generation demands efficient parallel inference techniques to manage their substantial computational cost. However, existing diffusion parallelism inference schemes…
View article: A Artifact-corrected artificial intelligence-LED intracoronary OCT analysis identifies plaque progression and vulnerability, drug efficacy and patient events
A Artifact-corrected artificial intelligence-LED intracoronary OCT analysis identifies plaque progression and vulnerability, drug efficacy and patient events Open
View article: RecGPT Technical Report
RecGPT Technical Report Open
Recommender systems are among the most impactful applications of artificial intelligence, serving as critical infrastructure connecting users, merchants, and platforms. However, most current industrial systems remain heavily reliant on his…
View article: Sustainable Factor Augmented Machine Learning Models for Crude Oil Return Forecasting
Sustainable Factor Augmented Machine Learning Models for Crude Oil Return Forecasting Open
The global crude oil market, known for its pronounced volatility and nonlinear dynamics, plays a pivotal role in shaping economic stability and informing investment strategies. Contrary to traditional research focused on price forecasting,…
View article: Investigation on the Properties of Alkali-Activated Industrial Solid Waste and Excavated-Soil-Based Controlled Low-Strength Materials
Investigation on the Properties of Alkali-Activated Industrial Solid Waste and Excavated-Soil-Based Controlled Low-Strength Materials Open
This study aims to address the challenge of backfill compaction in the confined spaces of municipal utility tunnel trenches and to develop an environmentally friendly, zero-cement-based backfill material. The research focuses on the excava…
View article: Time-o1: Time-Series Forecasting Needs Transformed Label Alignment
Time-o1: Time-Series Forecasting Needs Transformed Label Alignment Open
Training time-series forecast models presents unique challenges in designing effective learning objectives. Existing methods predominantly utilize the temporal mean squared error, which faces two critical challenges: (1) label autocorrelat…
View article: Review and recommendations for using artificial intelligence in intracoronary optical coherence tomography analysis
Review and recommendations for using artificial intelligence in intracoronary optical coherence tomography analysis Open
Artificial intelligence (AI) tools hold great promise for the rapid and accurate diagnosis of coronary artery disease (CAD) from intravascular optical coherent tomography (IVOCT) images. Numerous papers have been published describing AI-ba…
View article: TACO: Tackling Over-correction in Federated Learning with Tailored Adaptive Correction
TACO: Tackling Over-correction in Federated Learning with Tailored Adaptive Correction Open
Non-independent and identically distributed (Non-IID) data across edge clients have long posed significant challenges to federated learning (FL) training in edge computing environments. Prior works have proposed various methods to mitigate…
View article: Optical Differentiation and Edge Detection Based on Birefringence of Uniaxial Crystals
Optical Differentiation and Edge Detection Based on Birefringence of Uniaxial Crystals Open
Optical differential operations can directly extract edge information from images and have significant application potential in fields such as image processing and object recognition. In this work, we propose an optical spatial differentia…
View article: Think Before Recommend: Unleashing the Latent Reasoning Power for Sequential Recommendation
Think Before Recommend: Unleashing the Latent Reasoning Power for Sequential Recommendation Open
Sequential Recommendation (SeqRec) aims to predict the next item by capturing sequential patterns from users' historical interactions, playing a crucial role in many real-world recommender systems. However, existing approaches predominantl…
View article: Comparing between virtual reality based pre-clinical implantation training and traditional learning methods
Comparing between virtual reality based pre-clinical implantation training and traditional learning methods Open
Objective As dental implanting becomes an increasing demand among patients with tooth loss, an efficient and effective training for students is to be necessary. In this case, we anticipate the possible application of virtual reality (VR) t…
View article: Review and Recommendations for using Artificial Intelligence in Intracoronary Optical Coherence Tomography Analysis
Review and Recommendations for using Artificial Intelligence in Intracoronary Optical Coherence Tomography Analysis Open
Artificial intelligence (AI) methodologies hold great promise for the rapid and accurate diagnosis of coronary artery disease (CAD) from intravascular optical coherent tomography (IVOCT) images. Numerous papers have been published describi…
View article: Research on the Mechanism and Path of the Integration of Digital-Enabled Civic and Political Education in Colleges and Universities with Dual Innovation Education
Research on the Mechanism and Path of the Integration of Digital-Enabled Civic and Political Education in Colleges and Universities with Dual Innovation Education Open
Sorting out the relationship between civic and political education and dual-creation education makes the two fully integrated and can play a synergistic role in educating people. This paper carries out a preliminary analysis on the integra…
View article: Improving Retrospective Language Agents via Joint Policy Gradient Optimization
Improving Retrospective Language Agents via Joint Policy Gradient Optimization Open
View article: Spatiotemporal Neural Network with Attention Embeddings as Surrogate for Reservoir Automatic History Matching
Spatiotemporal Neural Network with Attention Embeddings as Surrogate for Reservoir Automatic History Matching Open
View article: BATseg: Boundary-aware Multiclass Spinal Cord Tumor Segmentation on 3D MRI Scans
BATseg: Boundary-aware Multiclass Spinal Cord Tumor Segmentation on 3D MRI Scans Open
Spinal cord tumors significantly contribute to neurological morbidity and mortality. Precise morphometric quantification, encompassing the size, location, and type of such tumors, holds promise for optimizing treatment planning strategies.…
View article: SEC-DT: Satellite Edge Computing Enabled Dynamic Data Transmission Based on GNN-Assisted MARL for Earth Observation Missions
SEC-DT: Satellite Edge Computing Enabled Dynamic Data Transmission Based on GNN-Assisted MARL for Earth Observation Missions Open
Recent advancements in low Earth orbit (LEO) satellite technology have facilitated a substantial increase in the number of Earth observation (EO) satellites launched. However, transmitting voluminous imagery generated by these EO satellite…
View article: FedReMa: Improving Personalized Federated Learning via Leveraging the Most Relevant Clients
FedReMa: Improving Personalized Federated Learning via Leveraging the Most Relevant Clients Open
Federated Learning (FL) is a distributed machine learning paradigm that achieves a globally robust model through decentralized computation and periodic model synthesis, primarily focusing on the global model's accuracy over aggregated data…
View article: FedReMa: Improving Personalized Federated Learning via Leveraging the Most Relevant Clients
FedReMa: Improving Personalized Federated Learning via Leveraging the Most Relevant Clients Open
Federated Learning (FL) is a distributed machine learning paradigm that achieves a globally robust model through decentralized computation and periodic model synthesis, primarily focusing on the global model’s accuracy over aggregated data…
View article: Model-in-the-Loop (MILO): Accelerating Multimodal AI Data Annotation with LLMs
Model-in-the-Loop (MILO): Accelerating Multimodal AI Data Annotation with LLMs Open
The growing demand for AI training data has transformed data annotation into a global industry, but traditional approaches relying on human annotators are often time-consuming, labor-intensive, and prone to inconsistent quality. We propose…
View article: Pluto and Charon: A Time and Memory Efficient Collaborative Edge AI Framework for Personal LLMs Fine-Tuning
Pluto and Charon: A Time and Memory Efficient Collaborative Edge AI Framework for Personal LLMs Fine-Tuning Open
Large language models (LLMs) have unlocked a plethora of powerful applications at the network edge, such as intelligent personal assistants. Data privacy and security concerns have prompted a shift towards edge-based fine-tuning of persona…
View article: Asteroid: Resource-Efficient Hybrid Pipeline Parallelism for Collaborative DNN Training on Heterogeneous Edge Devices
Asteroid: Resource-Efficient Hybrid Pipeline Parallelism for Collaborative DNN Training on Heterogeneous Edge Devices Open
On-device Deep Neural Network (DNN) training has been recognized as crucial for privacy-preserving machine learning at the edge. However, the intensive training workload and limited onboard computing resources pose significant challenges t…
View article: Alleviating All-to-All Communication for Deep Learning Recommendation Model Inference
Alleviating All-to-All Communication for Deep Learning Recommendation Model Inference Open
Massive DLRMs require large-scale multi-node systems for distributed training and inference, thus suffering from the all-to-all communication bottleneck. We propose an architecture, EmbedSwitch, that offloads the cache function of the embe…
View article: Pluto and Charon: A Time and Memory Efficient Collaborative Edge AI Framework for Personal LLMs Fine-tuning
Pluto and Charon: A Time and Memory Efficient Collaborative Edge AI Framework for Personal LLMs Fine-tuning Open
Large language models (LLMs) have unlocked a plethora of powerful applications at the network edge, such as intelligent personal assistants. Data privacy and security concerns have prompted a shift towards edge-based fine-tuning of persona…
View article: POSTER:In-network Model Inference for Distributed Systems via Programmable Switches
POSTER:In-network Model Inference for Distributed Systems via Programmable Switches Open
Model parallelism is crucial for accelerating distributed DNN inference. As a core component of distributed systems, programmable switches have demonstrated their ability in assisting both communication and partial computation. Exploiting …
View article: Edge Graph Intelligence: Reciprocally Empowering Edge Networks with Graph Intelligence
Edge Graph Intelligence: Reciprocally Empowering Edge Networks with Graph Intelligence Open
Recent years have witnessed a thriving growth of computing facilities connected at the network edge, cultivating edge networks as a fundamental infrastructure for supporting miscellaneous intelligent services.Meanwhile, Artificial Intellig…
View article: Identification of Multi-landscape and Cell Interactions in the Tumor Microenvironment through High-Coverage Single-Cell Sequencing
Identification of Multi-landscape and Cell Interactions in the Tumor Microenvironment through High-Coverage Single-Cell Sequencing Open
Single-cell RNA sequencing (scRNA-seq) is a widely used method for classifying cell types and states and revealing disease mechanisms. However, most contemporary scRNA-seq platforms fail to explore the multilandscape of RNA. Here, we desig…