Jie Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jie Yu has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Bin Ji
National university of defense technology, Changsha, China
5 shared publications - Jun Ma
National university of defense technology, Changsha, China
5 shared publications - Xiaodong Liu
National university of defense technology, Changsha, China
4 shared publications - Shangwen Wang
National university of defense technology, Changsha, China
3 shared publications - Shasha Li
National university of defense technology, Changsha, China
3 shared publications - Shezheng Song
College of Computer, National University of Defense Technology, Changsha, China
3 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 12010-12019
Beyond Token Stacking: Multi-scale Visual Modulation and Adaptive Fused-feature Calibration for MLLMs
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 12010-12019
Open AccessBeyond Token Stacking: Multi-scale Visual Modulation and Adaptive Fused-feature Calibration for MLLMs
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
AkiraRust: Re-thinking LLM-aided Rust Repair Using a Feedback-guided Thinking Switch
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessAkiraRust: Re-thinking LLM-aided Rust Repair Using a Feedback-guided Thinking Switch
- Open Access
- 30 June 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Software Engineering
- Vol. 3 (FSE) , 4323-4346
Three Heads Are Better Than One: A Multi-perspective Reasoning Framework for Enhanced Vulnerability Detection
Automated vulnerability detection is crucial for enhancing software security by identifying potential flaws that attackers could exploit, thereby reducing the reliance on labor-intensive manual code audits. Recent advancements have shifted towards leveraging large language models (LLMs) for vulnerability detection, with techniques like Vul-RAG and VulnSage demonstrating progress through structured prompting and external knowledge integration. However, these approaches typically rely on a single reasoning paradigm, limiting their ability to address the complex and diverse nature of real-world vulnerabilities. To overcome these limitations, we propose ReasonVul, a novel multi-perspective reasoning framework that harnesses cognitive synergy among three specialized LLM agents, each embodying a distinct reasoning mode. The framework begins with independent analyses of the source code, followed by a structured debate mechanism to resolve conflicts through iterative rebuttal and revision, ultimately converging on a collaborative judgment. Evaluated on the PrimeVul dataset, ReasonVul achieves a PairAcc of 40.00% and an F1-score of 72.52%, surpassing the best baseline by 81.24% in PairAcc. Further tests on the JITVUL dataset confirm its generalizability, with a PairAcc of 28.67%. Additionally, we analyzed 542 conflict cases and found that 389 were correctly resolved, highlighting the framework's ability to uncover hidden vulnerabilities through the error-correction mechanism driven by the debate. This work emphasizes the importance of multi-perspective reasoning and collaborative validation in achieving robust and comprehensive vulnerability detection in real-world software systems.
- 30 June 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Software Engineering
- Vol. 3 (FSE) , 4323-4346
Open AccessThree Heads Are Better Than One: A Multi-perspective Reasoning Framework for Enhanced Vulnerability Detection
Automated vulnerability detection is crucial for enhancing software security by identifying potential flaws that attackers could exploit, thereby reducing the reliance on labor-intensive manual code audits. Recent advancements have shifted towards leveraging large language models (LLMs) for vulnerability detection, with techniques like Vul-RAG and VulnSage demonstrating progress through structured prompting and external knowledge integration. However, these approaches typically rely on a single reasoning paradigm, limiting their ability to address the complex and diverse nature of real-world vulnerabilities. To overcome these limitations, we propose ReasonVul, a novel multi-perspective reasoning framework that harnesses cognitive synergy among three specialized LLM agents, each embodying a distinct reasoning mode. The framework begins with independent analyses of the source code, followed by a structured debate mechanism to resolve conflicts through iterative rebuttal and revision, ultimately converging on a collaborative judgment. Evaluated on the PrimeVul dataset, ReasonVul achieves a PairAcc of 40.00% and an F1-score of 72.52%, surpassing the best baseline by 81.24% in PairAcc. Further tests on the JITVUL dataset confirm its generalizability, with a PairAcc of 28.67%. Additionally, we analyzed 542 conflict cases and found that 389 were correctly resolved, highlighting the framework's ability to uncover hidden vulnerabilities through the error-correction mechanism driven by the debate. This work emphasizes the importance of multi-perspective reasoning and collaborative validation in achieving robust and comprehensive vulnerability detection in real-world software systems.
- Open Access
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 3847-3858
EMSEdit: Efficient Multi-Step Meta-Learning-based Model Editing
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 3847-3858
Open AccessEMSEdit: Efficient Multi-Step Meta-Learning-based Model Editing
- 3 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 37 (6) , 2658-2672
Evaluating Large Language Models on Named Entity Recognition
Large language models (LLMs) are popping up all over the place, and they have been gaining prominence due to their exceptional abilities in conducting various tasks. Although extensive LLM evaluation has been explored on natural language understanding tasks like text classification and sentiment analysis, evaluating LLMs on named entity recognition (NER) still remains under-explored. To fill this gap, we evaluate twenty-eight representative LLMs on thirteen datasets across five domains, whose parameters range from 3 billion to 175billion, from four perspectives, that is, supervised fine-tuning (SFT), parameter scales, hallucinations, and prompt designs. We propose an LLM-based NER framework (LLM-NER) for the evaluation, which consists of a Recognition phase and a Check phase. Specifically, the Check guides LLMs to examine the correctness of recognized entities, which is designed to mitigate hallucinations in the NER scenario. Qualitative and quantitative evaluation analyses demonstrate that in the NER scenario: 1) SFT empowers LLMs to understand and follow human instructions; 2) LLMs’ ability generally improves as their parameter scales consistently increase; 3) hallucinations exist in all evaluated LLMs, and guiding LLMs to check their outputs is a feasible way to alleviate hallucinations; and 4) all evaluated LLMs are sensitive to prompt designs. Based on the analyses, we highlight a number of promising directions for future study. Moreover, our evaluation shows high consistency with two LLM evaluation leaderboards, which evaluate LLMs on other tasks, demonstrating the rationality of our evaluation design.
- 3 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 37 (6) , 2658-2672
Evaluating Large Language Models on Named Entity Recognition
Large language models (LLMs) are popping up all over the place, and they have been gaining prominence due to their exceptional abilities in conducting various tasks. Although extensive LLM evaluation has been explored on natural language understanding tasks like text classification and sentiment analysis, evaluating LLMs on named entity recognition (NER) still remains under-explored. To fill this gap, we evaluate twenty-eight representative LLMs on thirteen datasets across five domains, whose parameters range from 3 billion to 175billion, from four perspectives, that is, supervised fine-tuning (SFT), parameter scales, hallucinations, and prompt designs. We propose an LLM-based NER framework (LLM-NER) for the evaluation, which consists of a Recognition phase and a Check phase. Specifically, the Check guides LLMs to examine the correctness of recognized entities, which is designed to mitigate hallucinations in the NER scenario. Qualitative and quantitative evaluation analyses demonstrate that in the NER scenario: 1) SFT empowers LLMs to understand and follow human instructions; 2) LLMs’ ability generally improves as their parameter scales consistently increase; 3) hallucinations exist in all evaluated LLMs, and guiding LLMs to check their outputs is a feasible way to alleviate hallucinations; and 4) all evaluated LLMs are sensitive to prompt designs. Based on the analyses, we highlight a number of promising directions for future study. Moreover, our evaluation shows high consistency with two LLM evaluation leaderboards, which evaluate LLMs on other tasks, demonstrating the rationality of our evaluation design.
- Open Access
- 16 December 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 22 (4) , 1-26
HotLD: A Workload-Aware Method for Global Code-Layout Optimization of Shared Libraries
Dynamic linking is an important technique in the process of software development. While dynamic linking can save memory and enhance maintainability, it also incurs performance overhead and hinders the application of profile-guided code layout optimization techniques to third-party libraries. Existing works (such as BOLT) can solve the problem by generating optimized versions of shared libraries for a given workload. Online PGO methods (such as OCOLOS) can further support on-the-fly code replacement to adapt to workload changes. These methods, however, are limited in several aspects: (1) hard to perform global optimization, (2) high memory consumption and performance overhead, and (3) limited usage scenarios. To address these issues, we propose HotLD , a workload-aware method for global code-layout optimization of shared libraries. HotLD can improve the performance of shared libraries while introducing limited performance and memory overhead. The core idea behind HotLD is to create dedicated code copies for each typical application workload during the offline phase, and perform global optimization according to workload characteristics. At runtime, HotLD monitors the running workload of the target program, then dynamically selects and links an appropriate code copy. We conducted experiments using real-world applications to evaluate the effectiveness and efficiency of HotLD . The results demonstrate that HotLD can improve their performance (up to 22.37%) with limited performance overhead (at the millisecond level). Compared with existing works, HotLD uses 1%–46% memory to support 2×–9× workloads.
- 16 December 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Architecture and Code Optimization
- Vol. 22 (4) , 1-26
Open AccessHotLD: A Workload-Aware Method for Global Code-Layout Optimization of Shared Libraries
Dynamic linking is an important technique in the process of software development. While dynamic linking can save memory and enhance maintainability, it also incurs performance overhead and hinders the application of profile-guided code layout optimization techniques to third-party libraries. Existing works (such as BOLT) can solve the problem by generating optimized versions of shared libraries for a given workload. Online PGO methods (such as OCOLOS) can further support on-the-fly code replacement to adapt to workload changes. These methods, however, are limited in several aspects: (1) hard to perform global optimization, (2) high memory consumption and performance overhead, and (3) limited usage scenarios. To address these issues, we propose HotLD , a workload-aware method for global code-layout optimization of shared libraries. HotLD can improve the performance of shared libraries while introducing limited performance and memory overhead. The core idea behind HotLD is to create dedicated code copies for each typical application workload during the offline phase, and perform global optimization according to workload characteristics. At runtime, HotLD monitors the running workload of the target program, then dynamically selects and links an appropriate code copy. We conducted experiments using real-world applications to evaluate the effectiveness and efficiency of HotLD . The results demonstrate that HotLD can improve their performance (up to 22.37%) with limited performance overhead (at the millisecond level). Compared with existing works, HotLD uses 1%–46% memory to support 2×–9× workloads.
- Open Access
- 12 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 132-139
SWSC: Compressing Large Language Models via Sharing Weights for Similar Channels
- 12 December 2025
- Published by Association for Computing Machinery (ACM)
- p. 132-139
Open AccessSWSC: Compressing Large Language Models via Sharing Weights for Similar Channels
- 24 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (24) , 51927-51950
A Survey of AI Inference Technologies for On-Device Systems
In recent years, artificial intelligence (AI) technologies represented by foundation models have experienced rapid development. Concurrently, on-device AI inference has become the primary approach for intelligent technology applications, offering advantages such as low latency, high security, and personalization. However, due to the limited resources of on-device systems, on-device AI inference faces new challenges, including improving computational efficiency, optimizing task parallelism, and model optimization. This survey addresses these challenges from a software and algorithmic perspective, focusing on three key areas—operator computation: explore methods to accelerate matrix multiplication and convolution, as well as techniques like operator fusion and vectorized computation; task inference: analyze heterogeneous and distributed computing, memory allocation, and energy-efficient tuning to improve the parallel execution and energy efficiency of inference tasks; and AI models: cover model compression, lookup table (LUT) quantization, and model architecture design to reduce computational complexity and storage requirements. By analyzing these areas, the survey aims to improve inference speed, reduce resource dependency, and provide insights into the future trends of on-device AI technology.
- 24 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (24) , 51927-51950
A Survey of AI Inference Technologies for On-Device Systems
In recent years, artificial intelligence (AI) technologies represented by foundation models have experienced rapid development. Concurrently, on-device AI inference has become the primary approach for intelligent technology applications, offering advantages such as low latency, high security, and personalization. However, due to the limited resources of on-device systems, on-device AI inference faces new challenges, including improving computational efficiency, optimizing task parallelism, and model optimization. This survey addresses these challenges from a software and algorithmic perspective, focusing on three key areas—operator computation: explore methods to accelerate matrix multiplication and convolution, as well as techniques like operator fusion and vectorized computation; task inference: analyze heterogeneous and distributed computing, memory allocation, and energy-efficient tuning to improve the parallel execution and energy efficiency of inference tasks; and AI models: cover model compression, lookup table (LUT) quantization, and model architecture design to reduce computational complexity and storage requirements. By analyzing these areas, the survey aims to improve inference speed, reduce resource dependency, and provide insights into the future trends of on-device AI technology.
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1553-1564
Identifying Knowledge Editing Types in Large Language Models
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1553-1564
Open AccessIdentifying Knowledge Editing Types in Large Language Models