Hairui Zhao

Orcid: 0009-0008-7081-5172

Affiliations:
  • Jilin University, Jilin, China


According to our database1, Hairui Zhao authored at least 17 papers between 2023 and 2026.

Collaborative distances:

Timeline

Legend:

Book  In proceedings  Article  PhD thesis  Dataset  Other 

Links

Online presence:

On csauthors.net:

Bibliography

2026
KVServe: Service-Aware KV Cache Compression for Communication-Efficient Disaggregated LLM Serving.
CoRR, May, 2026

TACO: Efficient Communication Compression of Intermediate Tensors for Scalable Tensor-Parallel LLM Training.
CoRR, April, 2026

A Fully GPU-Accelerated Framework for High-Performance Configuration Interaction Selection with Neural Network Quantum States.
CoRR, April, 2026

ENEC: A Lossless AI Model Compression Method Enabling Fast Inference on Ascend NPUs.
CoRR, April, 2026

COCCL: A Collective Communication Library Supporting Easy Integration and Configuration of Customized Compression for Scalable LLM Training.
Proceedings of the 31st ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, 2026

CCL-D: A High-Precision Diagnostic System for Slow and Hang Anomalies in Large-Scale Model Training.
Proceedings of the 31st ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, 2026

PRISM: An Efficient GPU-Based Lossy Compression Framework for Progressive Data Retrieval with Multi-Level Interpolation.
Proceedings of the 31st ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, 2026

2025
TurboFNO: High-Performance Fourier Neural Operator with Fused FFT-GEMM-iFFT on GPU.
CoRR, April, 2025

FT-Transformer: Resilient and Reliable Transformer with End-to-End Fault Tolerant Attention.
CoRR, April, 2025

Convergence-aware optimal checkpointing for exploratory deep learning training jobs.
Future Gener. Comput. Syst., 2025

Alleviating straggler impacts for data parallel deep learning with hybrid parameter update.
Future Gener. Comput. Syst., 2025

FlexPipe: Maximizing Training Efficiency for Transformer-based Models with Variable-Length Inputs.
Proceedings of the 2025 USENIX Annual Technical Conference, 2025

ArrayPipe: Introducing Job-Array Pipeline Parallelism for High Throughput Model Exploration.
Proceedings of the IEEE INFOCOM 2025, 2025

2024
Interference-aware opportunistic job placement for shared distributed deep learning clusters.
J. Parallel Distributed Comput., January, 2024

Visage: Visual-Aware Generation of Adversarial Examples in Black-Box for Text Classification.
Proceedings of the Natural Language Processing and Chinese Computing, 2024

EFNAS: Efficient Federated Neural Architecture Search Across AIoT Devices.
Proceedings of the International Joint Conference on Neural Networks, 2024

2023
ExplSched: Maximizing Deep Learning Cluster Efficiency for Exploratory Jobs.
Proceedings of the IEEE International Conference on Cluster Computing, 2023


  Loading...