Wencong Xiao

Alibaba Group (United States) — United States

Computer Science · Physical Sciences

13h-index1.2kcitations51works2.02yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Advanced Neural Network Applications(13), Cloud Computing and Resource Management(11), Stochastic Gradient Optimization Techniques(8), Parallel Computing and Optimization Techniques(7), Advanced Graph Neural Networks(6)

Publications51 total

Fast LLM Post-training via Decoupled and Fastest-of-N Speculation
ArXiv.org·2025Open Access
Preparation and performance study of phytic acid cytosine oxidized graphene hybrid modified multifunctional fabric
Colloids and Surfaces A Physicochemical and Engineering Aspects·2025
Mycroft: Tracing Dependencies in Collective Communication Towards Reliable LLM Training
2025
Robust LLM Training Infrastructure at ByteDance
2025Open Access
Deep Feature Decoupling Network for Ball Mill Load Signals
Machines·2025Open Access
Efficient Pre-Training of LLMs via Topology-Aware Communication Alignment on More Than 9600 GPUs
arXiv (Cornell University)·2025Open Access
Mycroft: Tracing Dependencies in Collective Communication Towards Reliable LLM Training
arXiv (Cornell University)·2025Open Access
Voyager: Input-Adaptive Algebraic Transformations for High-Performance Graph Neural Networks
2025Open Access
BootSeer: Analyzing and Mitigating Initialization Bottlenecks in Large-Scale LLM Training
arXiv (Cornell University)·2025Open Access
Data augmented offline deep reinforcement learning for stochastic dynamic power dispatch
International Journal of Electrical Power & Energy Systems·2025· 2 citedOpen Access
Polyphosphonitrile modified fabric: Preparation and the improvement of flame retardancy, hydrophobic and abrasion resistance
Progress in Organic Coatings·2025· 4 cited
Enhancing Large-Scale AI Training Efficiency: The C4 Solution for Real-Time Anomaly Detection and Communication Optimization
2025· 2 cited
Subdivision load identification of ball mill based on multi-domain feature extraction and UMAP-BOXGBoost
Signal Image and Video Processing·2025
Preparation and Performance Study of Phytic Acid Cytosine Oxidized Graphene Hybrid Modified Multifunctional Fabric
SSRN Electronic Journal·2025Open Access
Proactive Topology Optimization for Resilient Distribution Network Considering Robotaxi Coordinated Dispatch
SSRN Electronic Journal·2025Open Access
Towards Resilient Distribution System Via Microgrid Formulation: A Safe Hierarchical Hybrid Reinforcement Learning Approach
SSRN Electronic Journal·2025Open Access
Fine-Grained Structured Sparse Computing for FPGA-Based AI Inference
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems·2024· 1 cited
Rubick: Exploiting Job Reconfigurability for Deep Learning Cluster Scheduling
arXiv (Cornell University)·2024Open Access
Crux: GPU-Efficient Communication Scheduling for Deep Learning Training
2024· 23 cited
ElasticBatch: A Learning-Augmented Elastic Scheduling System for Batch Inference on MIG
IEEE Transactions on Parallel and Distributed Systems·2024· 2 cited
page 1 of 3Next →

Frequent Co-authors

Wei Lin(5), Xianyan Jia(5), Lintao Zhang(4), Chen Zhang(4), Myeongjae Jeon(4), Ming Wu(4), Lidong Zhou(4), Shiru Ren(4), Hailong Yang(4), Zhongzhi Luan(4), Depei Qian(4), Yong Li(4), Hanyu Zhao(4), Zuquan Song(4), Shuguang Wang(4), Shi‐Jie Cao(3), Zhuliang Yao(3), Lanshun Nie(3), Yunxin Liu(3), Shivaram Venkataraman(3)