Yuan Wen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yuan Wen has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Chunwei Xia
University of Leeds, Leeds, United Kingdom
2 shared publications - Huimin Cui
State Key Lab of Processors, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China, University of Chinese Academy of Sciences, Beijing, China, and XCORESIGMA CO.,LTD., Beijing, China
2 shared publications - Jiacheng Zhao
State Key Lab of Processors, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
2 shared publications - Xiaobing Feng
SKLP, Institute of Computing Technology, CAS; University of Chinese Academy of Sciences, Beijing, Beijing, China
2 shared publications - Zheng Wang
University of Leeds, Leeds, United Kingdom
2 shared publications Chao Jiang
Lenovo Research, China
1 shared publication
Latest Publications
- Open Access
- 14 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 2283-2297
When Grammar Guides the Attack: Uncovering Control-Plane Vulnerabilities in LLMs with Structured Output
- 14 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 2283-2297
Open AccessWhen Grammar Guides the Attack: Uncovering Control-Plane Vulnerabilities in LLMs with Structured Output
- Open Access
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 452-465
ParDiff: Efficiently Parallelizing Reverse-Mode Automatic Differentiation with Direct Indexing
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 452-465
Open AccessParDiff: Efficiently Parallelizing Reverse-Mode Automatic Differentiation with Direct Indexing
- Open Access
- 30 June 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 22 (2) , 1-23
Dynamic Power Management Through Multi-agent Deep Reinforcement Learning for Heterogeneous Systems
Power management and optimization play a significant role in modern computer systems, from battery-powered devices to servers running in data centers. Existing approaches for power capping fail to meet the requirements presented by dynamic workloads, and the situation becomes even more severe, given the divergent energy efficiency of workloads on heterogeneous hardware platforms. Adaptively optimizing energy consumption for dynamic workloads presents a great challenge to heterogeneous systems. To tackle this challenge, we present a machine learning based method to improve system-level power efficiency. We employ multi-agent deep reinforcement learning (MADRL) to automatically explore the relationship between long-term performance and the power budget for workloads of different types on classic CPU-GPU heterogeneous platforms. Our framework equips each device with an agent, enabling decentralized control over its power budget while maintaining centralized coordination to maximize the running time of applications within a power cap. We evaluate our approach against state-of-the-art methods on CPU-GPU platforms. Experimental results show that our method improves performance by an average of 8.5%. Additionally, our method is significantly more stable compared to the state-of-the-art heuristic approach.
- 30 June 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 22 (2) , 1-23
Open AccessDynamic Power Management Through Multi-agent Deep Reinforcement Learning for Heterogeneous Systems
Power management and optimization play a significant role in modern computer systems, from battery-powered devices to servers running in data centers. Existing approaches for power capping fail to meet the requirements presented by dynamic workloads, and the situation becomes even more severe, given the divergent energy efficiency of workloads on heterogeneous hardware platforms. Adaptively optimizing energy consumption for dynamic workloads presents a great challenge to heterogeneous systems. To tackle this challenge, we present a machine learning based method to improve system-level power efficiency. We employ multi-agent deep reinforcement learning (MADRL) to automatically explore the relationship between long-term performance and the power budget for workloads of different types on classic CPU-GPU heterogeneous platforms. Our framework equips each device with an agent, enabling decentralized control over its power budget while maintaining centralized coordination to maximize the running time of applications within a power cap. We evaluate our approach against state-of-the-art methods on CPU-GPU platforms. Experimental results show that our method improves performance by an average of 8.5%. Additionally, our method is significantly more stable compared to the state-of-the-art heuristic approach.
- Open Access
- 17 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 286-301
Optimizing Deep Learning Inference via Global Analysis and Tensor Expressions
Optimizing deep neural network (DNN) execution is important but becomes increasingly difficult as DNN complexity grows. Existing DNN compilers cannot effectively exploit optimization opportunities across operator boundaries, leaving room for improvement. To address this challenge, we present Souffle, an open-source compiler that optimizes DNN inference across operator boundaries. Souffle creates a global tensor dependency graph using tensor expressions, traces data flow and tensor information, and partitions the computation graph into subprograms based on dataflow analysis and resource constraints. Within a subprogram, Souffle performs local optimization via semantic-preserving transformations, finds an optimized program schedule, and improves instruction-level parallelism and data reuse. We evaluated Souffle using six representative DNN models on an NVIDIA A100 GPU. Experimental results show that Souffle consistently outperforms six state-of-the-art DNN optimizers by delivering a geometric mean speedup of up to 3.7× over TensorRT and 7.8× over Tensorflow XLA.
- 17 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 286-301
Open AccessOptimizing Deep Learning Inference via Global Analysis and Tensor Expressions
Optimizing deep neural network (DNN) execution is important but becomes increasingly difficult as DNN complexity grows. Existing DNN compilers cannot effectively exploit optimization opportunities across operator boundaries, leaving room for improvement. To address this challenge, we present Souffle, an open-source compiler that optimizes DNN inference across operator boundaries. Souffle creates a global tensor dependency graph using tensor expressions, traces data flow and tensor information, and partitions the computation graph into subprograms based on dataflow analysis and resource constraints. Within a subprogram, Souffle performs local optimization via semantic-preserving transformations, finds an optimized program schedule, and improves instruction-level parallelism and data reuse. We evaluated Souffle using six representative DNN models on an NVIDIA A100 GPU. Experimental results show that Souffle consistently outperforms six state-of-the-art DNN optimizers by delivering a geometric mean speedup of up to 3.7× over TensorRT and 7.8× over Tensorflow XLA.