Jingyu Wang
wangjingyu@bupt.edu.cn
Office: Room 511 Research Building
XiTuCheng Rd. #10
Beijing, China, 100876
Jingyu (Jeric) Wang [ 王敬宇 ] is Tenured Professor of School of Computer Science, Beijing University of Posts and Telecommunications (BUPT).
Changjiang Scholar (Ministry of Education of China).
Director, Network Intelligence Research Center (NIRC) in the State Key Laboratory of Networking and Switching Technology.
| Lead, Intelligent Everything (IE Corps, 万物有灵战队) | Team Homepage: https://nirc.top/. |
Published BUPT’s first-ever papers in premier global venues such as NSDI, ASPLOS, SIGMOD, EuroSys, and CSCW.
AAAI 2023 Distinguished Paper Award.
IEEE Systems Journal 2021 Best Paper Award.
multiple ESI Highly Cited Papers.
Intellectual Property: 5 authorized U.S. patents, 50+ authorized Chinese patents, and 10+ standardized contributions (3GPP, IETF, ETSI, CCSA).
Books: Intelligent Service Cloud Networks, Key Technologies for 6G Network On-demand Services, and Deep Reinforcement Learning in Network Intelligence.
Open Source: Lead contributor to OPNFV Compass (OpenStack community) and contributor to the Apache TVM machine learning compiler.
Competitions: Led teams to multiple championships in international competitions including AITrans、ACM MM、ACM MMSys、SemEval、ICCV Hands、Apollo and so on.
Research Projects & Industrial Impact:Professor Wang has served as the Principal Investigator (PI) for over 10 national and ministerial-level projects, including the National Key R&D Program, NSFC Key Projects, the National 973 Program, and son on.
He has also led over 20 industrialization projects for giants such as Huawei, China Mobile, State Grid, and Meituan.
His team has been recognized as an Innovation Team by the Ministry of Education and an Innovation Group by the NSFC.
His “Intelligent Network Management and Control System” has seen large-scale deployment across 15 provincial units, including China Mobile and China Unicom.
Publications & Academic Achievements:He has published over 300 high-level papers in top-tier (CCF-A) journals and conferences, including:
Journals: IEEE ToN, JSAC, TPDS, TDSC, TMC, TKDE, TIP, TSE, TSC, ACM TACO and so on.
Conferences: USENIX NSDI, ATC; ACM SIGCOMM, ASPLOS, SIGMOD, EuroSys, CSCW, EuroSys; IEEE INFOCOM; NeurIPS, ACL, CVPR and so on.
Key Awards:1*State Science and Technology Progress Award.
3*CIC Science and Technology Progress Awards.
2*Ministry of Education Science and Technology Progress Awards.
1*Beijing Science and Technology Progress Award.
Stanford’s World’s Top 2% Scientists.
Leadership in Professional Organizations:Fellow of the China Institute of Communications.
Chair of TG5 and TSC Member of 6GANA.
Vice Chair of AI Edge TG2.
Director/Board Member: International Society of Intelligent Networking and Systems (INSAI); Global Network Innovation Alliance (NIDA); Software Defined Interconnection Technology and Industry Alliance.
Committee Member: Computing Power Network and Digital Twin Committees of CIC; Computing Power Network Committee of the China Information Association.
Biography & Research Interests:With dual background in industry and academia, Prof. Wang focuses on the intersection of Communication Networks and Artificial Intelligence.
His research emphasizes understanding human-computer interaction and control decision-making, advancing the theoretical framework and engineering practice of the transition from the Internet of Everything (IoE) to Intent-Driven Intelligence (IDI).
He has pioneered a comprehensive theoretical and practical framework for intent-driven intelligence, effectively bridging the gap between ubiquitous computing networks and artificial intelligence. His research addresses the fundamental challenge of translating physical signals into high-level human intent within large-scale, heterogeneous environments. By integrating networking orchestration, high-performance system deployment, and deep semantic understanding, he has enabled the transition from basic connectivity to autonomous, intent-aware digital organisms, with his systems seeing scale application across national critical infrastructure and industrial sectors.
Representative Academic Contributions:1. Intent-Driven Intelligent Networking and Policy Verification
He developed an intent-driven architecture for autonomous network management, employing formal verification methods such as edge-predicates to achieve real-time policy validation under high-volume traffic conditions. His research on service-migration-driven configuration synthesis enables rapid network reconfiguration and service assurance, significantly enhancing the reliability and automation of large-scale communication systems.
2. Machine Learning Systems and Foundation Model Optimization
To overcome the barriers of deploying complex models on resource-constrained edge devices, he pioneered deep learning acceleration techniques using dynamic granularity zoning and decentralized learning architectures. He optimized large language model operations through attention-based cache compression and channel-wise quantization, effectively reducing hardware dependency for executing intelligent models across heterogeneous edge clusters.
3. Multimodal Time Series Analysis and Intelligent Operations
He established a unified multimodal paradigm for time series analysis by conceptualizing numerical data as discrete semantic tokens, enabling zero-shot forecasting and seamless cross-modal alignment. Furthermore, he enhanced the intelligence of system operations (AIOps) by leveraging large language models for log parsing, semantic association mining, and automated resolution generation for complex microservice failures.
4. Cognition-Guided Visual Semantics and Intent Insight
He introduced a cognition-guided visual analysis framework that incorporates explicit logical constraints and implicit prior knowledge to model complex relationships between objects and scenes. His work also clarifies the internal reasoning mechanisms of large language models, providing metrics to distinguish between structured reasoning and statistical shortcuts, which significantly improves the truthfulness and transparency of vision-language models.
5. High-Precision Pose Estimation and Natural Interaction Paradigms
In the field of high-precision pose estimation, he resolved long-standing challenges of self-occlusion and depth ambiguity by fusing multimodal representations such as images, point clouds, and 3D Gaussian splatting. His development of calibration-free multiview reconstruction and kinematic-driven mid-air handwriting recognition has defined new paradigms for natural, non-contact human-computer interaction in immersive spatial computing environments.
Teaching:He is teaching cutting-edge interdisciplinary courses such as “Intelligent Transportation and Autonomous Driving” as well as computer science courses like Software-Defined Network Technology and Next-Generation Internet Technology and Protocols.
news
| Jan 6, 2026 | One of our paper has been accepted by IEEE Transactions on Parallel and Distributed Systems (TPDS) [CCF A]. |
|---|---|
| Dec 8, 2025 | One of our paper has been accepted by ACM International Conference on Management of Data (SIGMOD) [CCF A]. |
| Dec 8, 2025 | Two of our papers have been accepted by IEEE Conference on Computer Communications (INFOCOM) [CCF A]. |
| Nov 21, 2025 | One of our paper has been accepted by IEEE Transactions on Dependable and Secure Computing (TDSC) [CCF A]. |
| Sep 20, 2025 | Four of our papers have been accepted by Annual Conference on Neural Information Processing Systems (NeurIPS) [CCF A]. |
| Jul 11, 2025 | Four of our papers have been accepted by ACM Multimedia (MM) [CCF A]. |
selected publications
Journal Articles
- ACM TACO二区 / Q2 / CCF ADeepZoning: Re-accelerate CNN Inference with Zoning Graph for Heterogeneous Edge ClusterACM Trans. Archit. Code Optim.(TACO), vol. 22, (1), pp. 10:1–10:26, 2025
- IEEE TMC一区 / Q1 / CCF AHierarchical Index Retrieval-Driven Wireless Network Intent Translation With LLMIEEE Trans. Mob. Comput.(TMC), vol. 24, (10), pp. 9837–9851, 2025
- IEEE ToN二区 / Q2 / CCF AFast and Scalable Data Plane Verification for Burst Updates With Edge-PredicateIEEE Trans. Netw.(ToN), vol. 33, (3), pp. 1279–1294, 2025
- IEEE TITS一区 / Q1 / CCF BFlight Trajectory Control With Network-Oriented Hierarchical Reinforcement Learning for UAVs-Assisted Data Time-Sensitive IoTIEEE Trans. Intell. Transp. Syst.(TITS), vol. 26, (5), pp. 6332–6345, 2025
- IEEE ToN 二区 / Q2 / CCF ADynamic Network Slice for Bursty Edge TrafficIEEE/ACM Trans. Netw. (ToN), vol. 32, (4), pp. 3142–3157, 2024
- IEEE ToN 二区 / Q2 / CCF AFast and Scalable ACL Policy Solving Under Complex Constraints With Graph Neural NetworksIEEE/ACM Trans. Netw.(ToN), vol. 32, (5), pp. 4175–4190, 2024
- IEEE ToN 二区 / Q2 / CCF AStanding on the Shoulders of Giants: Cross-Slice Federated Meta Learning for Resource Orchestration to Cold-Start SliceIEEE/ACM Trans. Netw. (ToN), vol. 31, (2), pp. 828–845, 2023
- IEEE TPDS 一区 / Q1 / CCF AMulti-SP Network Slicing Parallel Relieving Edge Network ConflictIEEE Trans. Parallel Distributed Syst.(TPDS), vol. 34, (11), pp. 2860–2875, 2023
- IEEE JSAC 一区 / Q1 / CCF AFollowing the Correct Direction: Renovating Sparsified SGD Towards Global Optimization in Distributed Edge LearningIEEE J. Sel. Areas Commun. (JSAC), vol. 40, (2), pp. 499–514, 2022
- IEEE TVT 二区 / Q2 / CCF BKnowledge-Driven Service Offloading Decision for Vehicular Edge Computing: A Deep Reinforcement Learning ApproachIEEE Trans. Veh. Technol.(TVT)[ESI Highly Cited], vol. 68, (5), pp. 4192–4203, 2019
Conference Articles
- AAAI 2025 CORE A* / CCF AChatTime: A Unified Multimodal Time Series Foundation Model Bridging Numerical and Textual DataIn AAAI-25, Sponsored by the Association for the Advancement of Artificial Intelligence, February 25 - March 4, 2025, Philadelphia, PA, USA, 2025
- ACM EuroSys 2025 CORE A* / CCF AAtlas: Towards Real-Time Verification in Large-Scale Networks via a Native Distributed ArchitectureIn Proceedings of the Twentieth European Conference on Computer Systems, EuroSys 2025 , Rotterdam, The Netherlands, 30 March 2025 - 3 April 2025, 2025
- IEEE INFOCOM 2025CORE A* / CCF ANetwork CoPilot: Intent-Driven Network Configuration Updating for Service GuaranteeIn IEEE INFOCOM 2025 - IEEE Conference on Computer Communications, London, United Kingdom, May 19-22, 2025, 2025
- USENIX ATC 2025CORE A* / CCF ANetKeeper: Enhancing Network Resilience with Autonomous Network Configuration Update on Traffic Patterns and AnomaliesIn Proceedings of the 2025 USENIX Annual Technical Conference, USENIX ATC 2025, Boston, MA, USA, July 7-9, 2025, 2025
- ACM ASPLOS 2024 CORE A* / CCF ANetRen: Service Migration-Driven Network Renascence with Synthesizing Updated ConfigurationIn Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3, ASPLOS 2024 , La Jolla, CA, USA, 27 April 2024- 1 May 2024, 2024
- NeurIPS 2024CORE A* / CCF AFM-Delta: Lossless Compression for Storing Massive Fine-tuned Foundation ModelsIn Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 2024, NeurIPS 2024 , Vancouver, BC, Canada, December 10 - 15, 2024, 2024
- USENIX NSDI 2024CORE A* / CCF AEPVerifier: Accelerating Update Storms Verification with Edge-PredicateIn 21st USENIX Symposium on Networked Systems Design and Implementation, NSDI 2024 , Santa Clara, CA, April 15-17, 2024, 2024
- AAAI 2023 CORE A* / CCF ATwo Heads Are Better than One: Image-Point Cloud Network for Depth-Based 3D Hand Pose EstimationIn Thirty-Seventh AAAI Conference on Artificial Intelligence, AAAI 2023 , Washington, DC, USA, February 7-14, 2023, 2023
- ACL 2020 CORE A* / CCF AAdversarial and Domain-Aware BERT for Cross-Domain Sentiment AnalysisIn Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, ACL 2020 , Online, July 5-10, 2020, 2020
- IEEE CVPR 2019 CORE A* / CCF AOICSR: Out-In-Channel Sparsity Regularization for Compact Deep Neural NetworksIn IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2019 , Long Beach, CA, USA, June 16-20, 2019, 2019