王智彬｜南京大学｜学者主页

2026

Scheduling Mixed RL Rollouts Beyond Prefix Locality

Zetao Hong, Song Yuan, Yuanhao Ding, Yibo Zhu, Daxin Jiang, Zhibin Wang, Chen Tian

arXiv (Cornell University)

2026

MoE Proxy Models for Low-Cost Failure Reproduction and Diagnosis in LLM RL Post-Training

Yikai Wang, Chuansai Zhou, Yuhang Zhou, Weiqiang Wu, Cong Wu, Yue Deng, Ben Feng, Mingming Zhu, Beirong Zhou, Zhibin Wang, Sheng Zhong, Chen Tian, Wangze Zhang

arXiv (Cornell University)

DOI

2026

Adaptive Matrix Multiplication for Dynamic Shapes on Ascend NPUs

Yuhang Zhou, Jiang Peng, Qianyu Jiang, Zhibin Wang, Xinghui Tian, Jianwei Zhou, Songxiang Zhu, Jingyi Zhang, Junsong Wang, Chen Tian

arXiv preprint

2026

TIDE-MC: Two-Sided Interpolative Decomposition for Billion-Scale GPU Matrix Completion

Chengying Huan, Yubo Wang, Pinhuan Wang, Lizheng Chen, Jie Zhang, Fangxin Liu, Qing Wang, Ruixuan Liu, Shaonan Ma, Mingxing Zhang, Zhibin Wang, Rong Gu, Guihai Chen, Chen Tian

arXiv preprint

2026

SpecLA: Efficient Speculative Decoding for Linear-Attention Models

Zhibin Wang, Xuying Han, Zhaohua Yang, Fuliang Liu, Xue Li, Rong Gu, Sheng Zhong, Chen Tian

arXiv preprint

2026

LOCAL: Enabling Learning On-device Contiguously for Agent LLMs

Xinxin Liu, Jiaxin Li, Zibo Wang, Yun Ji, Zhangqi Zhu, Qing Hu, Zhibin Wang, Rong Gu, Sheng Zhong, Chen Tian

arXiv preprint

2026

DART: Diffusion-Inspired Speculative Decoding for Fast LLM Inference

Fuliang Liu, Xue Li, Ketai Zhao, Yinxi Gao, Ziyan Zhou, Zhonghui Zhang, Zhibin Wang, Wanchun Dou, Sheng Zhong, Chen Tian

arXiv preprint

2026

TreeSpark: Learning to Draft the Trees You Verify for Speculative Decoding

Jihao Zhao, Qianyu Jiang, Shuaishuai Zu, Pengfei Li, Guanghuan Fang, Ben Feng, Biao Qin, Zhibin Wang, Baochuan Yang

arXiv preprint

2026

SSV: Sparse Speculative Verification for Efficient LLM Inference

Zhibin Wang, Ziyu Zhong, Nuo Shen, Yuhang Zhou, Rong Gu, Sheng Zhong

arXiv preprint

2026

When Parallel Drafter Meets Parallel Speculative Decoding

Fuliang Liu, Xue Li, Kun Qian, Zhibin Wang, Wanchun Dou, Wenyuan Yu, Chen Tian

arXiv preprint

2025

SmartSwap: Swap-Based Memory Optimization for LLM Training under Varying Operator Sequences

Wang, Zibo, Yuhang Zhou, Zhibin Wang, Shipeng Li, Xinjing Huang, Cai, Chendong, Mu, Bingxu, Yuqing Sun, Zhiheng Hu, She, Bin, Shu You, Guo Hua Fang, Rong Gu, Wanchun Dou, Guihai Chen, Chen Tian

arXiv (Cornell University)

DOI

2025

Chordless Structure: A Pathway to Simple and Expressive GNNs

Hongxu Pan, Shuxian Hu, Mo Zhou, Zhibin Wang, Rong Gu, Chen Tian, Kun Yang, Sheng Zhong

arXiv preprint

2025

Echo: Efficient Co-Scheduling of Hybrid Online-Offline Tasks for Large Language Model Serving

Zhibin Wang, Shipeng Li, Xue Li, Yuhang Zhou, Zhonghui Zhang, Zibo Wang, Rong Gu, Chen Tian, Kun Yang, Sheng Zhong

arXiv preprint

2025

STAR: Decode-Phase Rescheduling for LLM Inference

Zhibin Wang, Zetao Hong, Xue Li, Zibo Wang, Shipeng Li, Qingkai Meng, Qing Wang, Chengying Huan, Rong Gu, Sheng Zhong, Chen Tian

arXiv preprint

DOI

2025

Bingo: Radix-based Bias Factorization for Random Walk on Dynamic Graphs

Pinhuan Wang, Chengying Huan, Zhibin Wang, Chen Tian, Yuede Ji, Hang Liu

arXiv preprint

DOI

2025

Accelerating Mixture-of-Experts Inference by Hiding Offloading Latency with Speculative Decoding

Zhibin Wang, Zhonghui Zhang, Yuhang Zhou, Zibo Wang, Mo Zhou, Peng Jiang, Weilin Cai, Chengying Huan, Rong Gu, Sheng Zhong, Chen Tian

arXiv preprint

2025

GMoPE:A Prompt-Expert Mixture Framework for Graph Foundation Models

Zhibin Wang, Zhixing Zhang, Shuqi Wang, Xuanting Xie, Zhao Kang

arXiv preprint

2025

CoDec: Prefix-Shared Decoding Kernel for LLMs

Zhibin Wang, Rui Ning, Chao Fang, Zhonghui Zhang, Xi Lin, Shaobo Ma, Mo Zhou, Xue Li, Zhongfeng Wang, Chengying Huan, Rong Gu, Kun Yang, Guihai Chen, Sheng Zhong, Chen Tian

arXiv preprint

DOI

2024

Revisiting Service Level Objectives and System Level Metrics in Large Language Model Serving

Zhibin Wang, Shipeng Li, Yuhang Zhou, Xue Li, Zhonghui Zhang, Nguyen Cam-Tu, Rong Gu, Chen Tian, Guihai Chen, Sheng Zhong

arXiv preprint

2023

Unicron: Economizing Self-Healing LLM Training at Scale

Tao He, Li, Xue, Zhibin Wang, Qian, Kun, Jing-Bo Xu, Wenyuan Yu, Jingren Zhou

arXiv (Cornell University)

DOI

2023

GraphAr: An Efficient Storage Scheme for Graph Data in Data Lakes

Xue Li, Weibin Zeng, Zhibin Wang, Diwen Zhu, Jingbo Xu, Wenyuan Yu, Jingren Zhou

arXiv preprint

2021

TRUST: Triangle Counting Reloaded on GPUs

Santosh Pandey, Zhibin Wang, Sheng Zhong, Chen Tian, Bolong Zheng, Xiaoye Li, Lingda Li, Adolfy Hoisie, Caiwen Ding, Dong Li, Hang Liu

arXiv preprint

DOI

王智彬

精选作品

期刊论文

大语言模型服务系统服务级目标和系统级指标优化

星脉3.0：面向MoE模型训推一体的大规模高性能网络设计与实践

Resonating with RoPE: Spectral Quantization for High-Fidelity Key Cache Compression

Quartet: A Query Aware Database Adaptive Compilation Decision System

SMOG: Accelerating Subgraph Matching on GPUs

预印本

Scheduling Mixed RL Rollouts Beyond Prefix Locality

MoE Proxy Models for Low-Cost Failure Reproduction and Diagnosis in LLM RL Post-Training

Adaptive Matrix Multiplication for Dynamic Shapes on Ascend NPUs

TIDE-MC: Two-Sided Interpolative Decomposition for Billion-Scale GPU Matrix Completion

SpecLA: Efficient Speculative Decoding for Linear-Attention Models

LOCAL: Enabling Learning On-device Contiguously for Agent LLMs

DART: Diffusion-Inspired Speculative Decoding for Fast LLM Inference

TreeSpark: Learning to Draft the Trees You Verify for Speculative Decoding

SSV: Sparse Speculative Verification for Efficient LLM Inference

When Parallel Drafter Meets Parallel Speculative Decoding

SmartSwap: Swap-Based Memory Optimization for LLM Training under Varying Operator Sequences

Chordless Structure: A Pathway to Simple and Expressive GNNs

Echo: Efficient Co-Scheduling of Hybrid Online-Offline Tasks for Large Language Model Serving

STAR: Decode-Phase Rescheduling for LLM Inference

Bingo: Radix-based Bias Factorization for Random Walk on Dynamic Graphs

Accelerating Mixture-of-Experts Inference by Hiding Offloading Latency with Speculative Decoding

GMoPE:A Prompt-Expert Mixture Framework for Graph Foundation Models

CoDec: Prefix-Shared Decoding Kernel for LLMs

Revisiting Service Level Objectives and System Level Metrics in Large Language Model Serving

Unicron: Economizing Self-Healing LLM Training at Scale

GraphAr: An Efficient Storage Scheme for Graph Data in Data Lakes

TRUST: Triangle Counting Reloaded on GPUs

研究方向