Yue Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yue Zhang has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Shicheng Zhou
College of Electronic Engineering, National University of Defense Technology, Hefei, China
7 shared publications - Jingju Liu
College of Electronic Engineering, National University of Defense Technology, Hefei, China
4 shared publications - Yuliang Lu
National University of Defense Technology, Hefei, China
4 shared publications - Dongdong Hou
College of Electronic Engineering, National University of Defense Technology, Hefei 230037, China
3 shared publications - Jiahai Yang
Tsinghua University, Beijing, China
3 shared publications - Jingju Liu
College of Electronic Engineering, National University of Defense Technology, Hefei 230037, China
2 shared publications
Latest Publications
- Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 996-1006
KnowFC: Navigating Knowledge Conflicts in Large Language Model-based Fact-Checking
When fact-checking methods based on large language models (LLMs) use external evidence to validate claims, knowledge conflicts often arise. These conflicts typically stem from inconsistencies between the external evidence and LLMs' internal pre-existing knowledge. Such an inconsistency could lead LLMs to draw incorrect answers when validating claims, especially when they are overly confident in their internal incorrect knowledge. Previous works on LLM-based fact-checking have overlooked this issue. This paper, for the first time, proposes a framework (namely KnowFC) to navigate this issue. Our key insight is dividing and adaptively utilizing the knowledge that LLMs know and do not know, thereby avoiding conflicts while enhancing the correctness and efficiency of fact-checking. Specifically, in KnowFC, we propose an adaptive retrieval method, where we train an LLM using a reinforcement learning algorithm coupled with the Dunning-Kruger effect-inspired reward mechanism to identify its knowledge boundaries through confidence calibration, thereby realizing adaptive evidence retrieval. Besides, we propose a reliable and debiased fact verification method, where we organize and construct reasoning graphs using retrieved evidence to verify claims, followed by a causal intervention method using causal mediation analysis to mitigate internal knowledge interference. Experimental results on both FEVEROUS and AVeriTeC datasets show that our method outperforms baseline methods in terms of accuracy and F1 score, while also improving fact-checking efficiency.
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 996-1006
Open AccessKnowFC: Navigating Knowledge Conflicts in Large Language Model-based Fact-Checking
When fact-checking methods based on large language models (LLMs) use external evidence to validate claims, knowledge conflicts often arise. These conflicts typically stem from inconsistencies between the external evidence and LLMs' internal pre-existing knowledge. Such an inconsistency could lead LLMs to draw incorrect answers when validating claims, especially when they are overly confident in their internal incorrect knowledge. Previous works on LLM-based fact-checking have overlooked this issue. This paper, for the first time, proposes a framework (namely KnowFC) to navigate this issue. Our key insight is dividing and adaptively utilizing the knowledge that LLMs know and do not know, thereby avoiding conflicts while enhancing the correctness and efficiency of fact-checking. Specifically, in KnowFC, we propose an adaptive retrieval method, where we train an LLM using a reinforcement learning algorithm coupled with the Dunning-Kruger effect-inspired reward mechanism to identify its knowledge boundaries through confidence calibration, thereby realizing adaptive evidence retrieval. Besides, we propose a reliable and debiased fact verification method, where we organize and construct reasoning graphs using retrieved evidence to verify claims, followed by a causal intervention method using causal mediation analysis to mitigate internal knowledge interference. Experimental results on both FEVEROUS and AVeriTeC datasets show that our method outperforms baseline methods in terms of accuracy and F1 score, while also improving fact-checking efficiency.
- 12 December 2025
- Published by Zhejiang University Press in Frontiers of Information Technology & Electronic Engineering
- Vol. 26 (12) , 2511-2528
Mind the Gap: towards generalizable autonomous penetration testing via domain randomization and meta-reinforcement learning
With the increasing number of vulnerabilities exposed on the Internet, autonomous penetration testing (pentesting) has emerged as a promising research area
- 12 December 2025
- Published by Zhejiang University Press in Frontiers of Information Technology & Electronic Engineering
- Vol. 26 (12) , 2511-2528
Mind the Gap: towards generalizable autonomous penetration testing via domain randomization and meta-reinforcement learning
With the increasing number of vulnerabilities exposed on the Internet, autonomous penetration testing (pentesting) has emerged as a promising research area
- 11 November 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 300
Autonomous penetration testing using reinforcement learning: A review and perspectives
- 11 November 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 300
Autonomous penetration testing using reinforcement learning: A review and perspectives
- 8 May 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 285
SCRIPT: A Scalable Continual Reinforcement Learning Framework for Autonomous Penetration Testing
- 8 May 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 285
SCRIPT: A Scalable Continual Reinforcement Learning Framework for Autonomous Penetration Testing
- 16 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (3) , 2443-2459
APRIL: Towards Scalable and Transferable Autonomous Penetration Testing in Large Action Space via Action Embedding
Penetration testing (pentesting) assesses cybersecurity through simulated attacks, while the conventional manual-based method is costly, time-consuming, and personnelconstrained. Reinforcement learning (RL) provides an agentenvironment interaction learning paradigm, making it a promising way for autonomous pentesting. However, agents' scalability in large action spaces and policy transferability across scenarios limit the applicability of RL-based autonomous pentesting. To address these challenges, we present a novel autonomous pentesting framework based on reinforcement learning (namely APRIL) to train agents that are scalable and transferable in large action spaces. In APRIL, we construct realistic, bounded, host-level state space via embedding techniques to avoid the complexities of dealing with unbounded network-level information. We employ semantic correlations between pentesting actions as prior knowledge to represent discrete action space into a continuous and semantically meaningful embedding space. Agents are then trained to reason over actions within the action embedding space, where two key methods are applied: an upper-confidence bound-based action refinement method to encourage efficient exploration, and a distance-aware loss to improve learning efficiency and generalization performance. We conduct experiments in simulated scenarios constructed based on virtualized vulnerable environments. The results demonstrate APRIL's scalability in large action spaces and its ability to facilitate policy transfer across diverse scenarios.
- 16 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 22 (3) , 2443-2459
APRIL: Towards Scalable and Transferable Autonomous Penetration Testing in Large Action Space via Action Embedding
Penetration testing (pentesting) assesses cybersecurity through simulated attacks, while the conventional manual-based method is costly, time-consuming, and personnelconstrained. Reinforcement learning (RL) provides an agentenvironment interaction learning paradigm, making it a promising way for autonomous pentesting. However, agents' scalability in large action spaces and policy transferability across scenarios limit the applicability of RL-based autonomous pentesting. To address these challenges, we present a novel autonomous pentesting framework based on reinforcement learning (namely APRIL) to train agents that are scalable and transferable in large action spaces. In APRIL, we construct realistic, bounded, host-level state space via embedding techniques to avoid the complexities of dealing with unbounded network-level information. We employ semantic correlations between pentesting actions as prior knowledge to represent discrete action space into a continuous and semantically meaningful embedding space. Agents are then trained to reason over actions within the action embedding space, where two key methods are applied: an upper-confidence bound-based action refinement method to encourage efficient exploration, and a distance-aware loss to improve learning efficiency and generalization performance. We conduct experiments in simulated scenarios constructed based on virtualized vulnerable environments. The results demonstrate APRIL's scalability in large action spaces and its ability to facilitate policy transfer across diverse scenarios.
- Open Access
- 14 October 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (20) , 10339
Improved Deep Recurrent Q-Network of POMDPs for Automated Penetration Testing
With the development of technology, people’s daily lives are closely related to networks. The importance of cybersecurity protection draws global attention. Automated penetration testing is the novel method to protect the security of networks, which enhances efficiency and reduces costs compared with traditional manual penetration testing. Previous studies have provided many ways to obtain a better policy for penetration testing paths, but many studies are based on ideal penetration testing scenarios. In order to find potential vulnerabilities from the perspective of hackers in the real world, this paper models the process of black-box penetration testing as a Partially Observed Markov Decision Process (POMDP). In addition, we propose a new algorithm named , which is applied to the automated black-box penetration testing. In the POMDP model, an agent interacts with a network environment to choose a better policy without insider information about the target network, except for the start points. To handle this problem, we utilize a Long Short-Term Memory (LSTM) structure empowering agent to make decisions based on historical memory. In addition, this paper enhances the current algorithm using the structure of the neural network, the calculation method of the Q-value, and adding noise parameters to the neural network to advance the generalization and efficiency of this algorithm. In the last section, we conduct comparison experiments of the algorithm and other recent state-of-the-art (SOTA) algorithms. The experimental results vividly show that this novel algorithm is able to find a greater attack-path strategy for all vulnerable hosts in the automated black-box penetration testing. Additionally, the generalization and robustness of this algorithm are far superior to other SOTA algorithms in different size simulation scenarios based on the CyberBattleSim simulation developed by Microsoft.- 14 October 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (20) , 10339
Open AccessImproved Deep Recurrent Q-Network of POMDPs for Automated Penetration Testing
With the development of technology, people’s daily lives are closely related to networks. The importance of cybersecurity protection draws global attention. Automated penetration testing is the novel method to protect the security of networks, which enhances efficiency and reduces costs compared with traditional manual penetration testing. Previous studies have provided many ways to obtain a better policy for penetration testing paths, but many studies are based on ideal penetration testing scenarios. In order to find potential vulnerabilities from the perspective of hackers in the real world, this paper models the process of black-box penetration testing as a Partially Observed Markov Decision Process (POMDP). In addition, we propose a new algorithm named , which is applied to the automated black-box penetration testing. In the POMDP model, an agent interacts with a network environment to choose a better policy without insider information about the target network, except for the start points. To handle this problem, we utilize a Long Short-Term Memory (LSTM) structure empowering agent to make decisions based on historical memory. In addition, this paper enhances the current algorithm using the structure of the neural network, the calculation method of the Q-value, and adding noise parameters to the neural network to advance the generalization and efficiency of this algorithm. In the last section, we conduct comparison experiments of the algorithm and other recent state-of-the-art (SOTA) algorithms. The experimental results vividly show that this novel algorithm is able to find a greater attack-path strategy for all vulnerable hosts in the automated black-box penetration testing. Additionally, the generalization and robustness of this algorithm are far superior to other SOTA algorithms in different size simulation scenarios based on the CyberBattleSim simulation developed by Microsoft. - Open Access
- 23 September 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (19) , 8823
Autonomous Penetration Testing Based on Improved Deep Q-Network
Penetration testing is an effective way to test and evaluate cybersecurity by simulating a cyberattack. However, the traditional methods deeply rely on domain expert knowledge, which requires prohibitive labor and time costs. Autonomous penetration testing is a more efficient and intelligent way to solve this problem. In this paper, we model penetration testing as a Markov decision process problem and use reinforcement learning technology for autonomous penetration testing in large scale networks. We propose an improved deep Q-network (DQN) named NDSPI-DQN to address the sparse reward problem and large action space problem in large-scale scenarios. First, we reasonably integrate five extensions to DQN, including noisy nets, soft Q-learning, dueling architectures, prioritized experience replay, and intrinsic curiosity model to improve the exploration efficiency. Second, we decouple the action and split the estimators of the neural network to calculate two elements of action separately, so as to decrease the action space. Finally, the performance of algorithms is investigated in a range of scenarios. The experiment results demonstrate that our methods have better convergence and scaling performance.
- 23 September 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (19) , 8823
Open AccessAutonomous Penetration Testing Based on Improved Deep Q-Network
Penetration testing is an effective way to test and evaluate cybersecurity by simulating a cyberattack. However, the traditional methods deeply rely on domain expert knowledge, which requires prohibitive labor and time costs. Autonomous penetration testing is a more efficient and intelligent way to solve this problem. In this paper, we model penetration testing as a Markov decision process problem and use reinforcement learning technology for autonomous penetration testing in large scale networks. We propose an improved deep Q-network (DQN) named NDSPI-DQN to address the sparse reward problem and large action space problem in large-scale scenarios. First, we reasonably integrate five extensions to DQN, including noisy nets, soft Q-learning, dueling architectures, prioritized experience replay, and intrinsic curiosity model to improve the exploration efficiency. Second, we decouple the action and split the estimators of the neural network to calculate two elements of action separately, so as to decrease the action space. Finally, the performance of algorithms is investigated in a range of scenarios. The experiment results demonstrate that our methods have better convergence and scaling performance.