Su Wang
Research Subjects
Su Wang has a total of 35 co-authors. Here are the top co-authors by number of shared publications:
- Jiahai Yang
Tsinghua University, Beijing, China
5 shared publications - Zhiliang Wang
Tsinghua University, Beijing, China
5 shared publications - Dongqi Han
Faculty of Beijing University of Posts and Telecommunications, Beijing, China
4 shared publications - Xingang Shi
Tsinghua University, Beijing, Beijing, China and Zhongguancun Laboratory, Beijing, Beijing, China
4 shared publications - Han Zhang
Tsinghua University, Beijing, Beijing, China and Zhongguancun Laboratory, Beijing, Beijing, China
3 shared publications - Hongbin Sun
Zhongguancun Laboratory, Beijing, China
3 shared publications
Latest Publications
- 19 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. PP (08908044) , 1-8
The Evolution of Provenance-Based Intrusion Detection Systems: A Review of Approaches, Limitations, and Future Directions
The provenance graph, due to its ability to fully preserve historical information within a host, has been extensively studied and applied in the field of host intrusion detection. Such research is collectively referred to as provenance-based intrusion detection systems (PIDSes). However, recent surveys on PIDSes face limitations such as outdated literature and insufficiently comprehensive classifications. To systematically summarize the evolutionary trajectory and experimental evaluation framework of PIDSes, this paper conducts a literature review from the perspective of PIDSes evolution, proposing a more refined framework for host threat detection. For each fundamental step within the framework, representative PIDSes are reviewed and introduced. Additionally, we qualitatively analyze state-of-the-art (SOTA) PIDSes from the perspective of experimental evaluation, summarize seven key findings derived from experimental evaluations, and provide an analysis of future research directions addressing some existing limitations in the field of PIDSes.
- 19 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. PP (08908044) , 1-8
The Evolution of Provenance-Based Intrusion Detection Systems: A Review of Approaches, Limitations, and Future Directions
The provenance graph, due to its ability to fully preserve historical information within a host, has been extensively studied and applied in the field of host intrusion detection. Such research is collectively referred to as provenance-based intrusion detection systems (PIDSes). However, recent surveys on PIDSes face limitations such as outdated literature and insufficiently comprehensive classifications. To systematically summarize the evolutionary trajectory and experimental evaluation framework of PIDSes, this paper conducts a literature review from the perspective of PIDSes evolution, proposing a more refined framework for host threat detection. For each fundamental step within the framework, representative PIDSes are reviewed and introduced. Additionally, we qualitatively analyze state-of-the-art (SOTA) PIDSes from the perspective of experimental evaluation, summarize seven key findings derived from experimental evaluations, and provide an analysis of future research directions addressing some existing limitations in the field of PIDSes.
- 11 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 7317-7332
End-to-End Attack Scene Reconstruction in a Host With Rules and Anomaly-Based Detection Models
Critical devices on the Internet are frequently targeted by skilled and advanced network attackers. These attackers often orchestrate complex and persistent intrusion campaigns, which involve multiple stages of attacks. In the context of host-based threat detection, the reconstruction of the entire attack scenario is crucial for tracing threats and fixing system vulnerabilities. Prior anomaly-based studies lack the capability to interpret the attack scenario, while rule-based approaches struggle with detecting novel attack patterns. We introduce eaGle, an end-to-end framework that takes original host-based data as input and reconstructs the potential attack scenario as output. It leverages an anomaly-based algorithm and a fine-grained misuse detection module to assign anomalous scores to host data, constructs the potential attack scenario using a novel anomalous subtree detection algorithm, and generates the interpretable attack scenario graph through a coarse-grained rule matching method. We assess the performance of eaGle using three attack scenarios from the DARPA TC dataset and three deployment scenarios. The results demonstrate that eaGle can effectively uncover the hidden attack scenario within the host data and outperforms three state-of-the-art attack scenario reconstruction systems.
- 11 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 7317-7332
End-to-End Attack Scene Reconstruction in a Host With Rules and Anomaly-Based Detection Models
Critical devices on the Internet are frequently targeted by skilled and advanced network attackers. These attackers often orchestrate complex and persistent intrusion campaigns, which involve multiple stages of attacks. In the context of host-based threat detection, the reconstruction of the entire attack scenario is crucial for tracing threats and fixing system vulnerabilities. Prior anomaly-based studies lack the capability to interpret the attack scenario, while rule-based approaches struggle with detecting novel attack patterns. We introduce eaGle, an end-to-end framework that takes original host-based data as input and reconstructs the potential attack scenario as output. It leverages an anomaly-based algorithm and a fine-grained misuse detection module to assign anomalous scores to host data, constructs the potential attack scenario using a novel anomalous subtree detection algorithm, and generates the interpretable attack scenario graph through a coarse-grained rule matching method. We assess the performance of eaGle using three attack scenarios from the DARPA TC dataset and three deployment scenarios. The results demonstrate that eaGle can effectively uncover the hidden attack scenario within the host data and outperforms three state-of-the-art attack scenario reconstruction systems.
- Open Access
- 2 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 4509-4523
Rules Refine the Riddle: Global Explanation for Deep Learning-Based Anomaly Detection in Security Applications
Deep learning (DL) based anomaly detection has shown great promise in the field of security due to its remarkable performance in various tasks. However, the issue of poor interpretability in DL models has significantly impeded their deployment in practical security applications. Despite the progress made in existing studies on DL explanations, the majority of them focus on providing local explanations for individual samples, neglecting the global understanding of the model knowledge. Furthermore, most explanations for supervised models fail to apply to anomaly detection due to their different learning mechanisms. In this work, we address the gap in the existing research by proposing GEAD, a novel global explanation for DL-based anomaly detection, to extract high-fidelity rules from DL models. We apply GEAD to two security applications, network intrusion detection and system log anomaly detection, and demonstrate the efficacy with three usages: comparing model knowledge with expert knowledge, identifying knowledge discrepancies between models, and combining model and expert knowledge. We provide several case studies to showcase how GEAD can significantly enhance existing anomaly detection systems. Moreover, we provide a real-world deployment in a SCADA system to showcase the potential in practice. Some important insights are drawn to help the community understand and improve anomaly detection systems in security.
- 2 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 4509-4523
Open AccessRules Refine the Riddle: Global Explanation for Deep Learning-Based Anomaly Detection in Security Applications
Deep learning (DL) based anomaly detection has shown great promise in the field of security due to its remarkable performance in various tasks. However, the issue of poor interpretability in DL models has significantly impeded their deployment in practical security applications. Despite the progress made in existing studies on DL explanations, the majority of them focus on providing local explanations for individual samples, neglecting the global understanding of the model knowledge. Furthermore, most explanations for supervised models fail to apply to anomaly detection due to their different learning mechanisms. In this work, we address the gap in the existing research by proposing GEAD, a novel global explanation for DL-based anomaly detection, to extract high-fidelity rules from DL models. We apply GEAD to two security applications, network intrusion detection and system log anomaly detection, and demonstrate the efficacy with three usages: comparing model knowledge with expert knowledge, identifying knowledge discrepancies between models, and combining model and expert knowledge. We provide several case studies to showcase how GEAD can significantly enhance existing anomaly detection systems. Moreover, we provide a real-world deployment in a SCADA system to showcase the potential in practice. Some important insights are drawn to help the community understand and improve anomaly detection systems in security.
- Open Access
- 30 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 263-277
AudiTrim: A Real-time, General, Efficient, and Low-overhead Data Compaction System for Intrusion Detection
Recently enterprises and governments face escalating APT attacks, leading to significant economic losses. APT attacks often persist for extended periods, necessitating the storage of extensive audit logs for effective detection. To reduce data storage overhead, enterprises commonly adopt compression strategies. However, efficient compression strategies may introduce additional query overhead. Existing approaches propose data reduction algorithms, but these methods can compromise data integrity, rendering current attack investigation and anomaly-based intrusion detection ineffective. To address these difficulties, we present AudiTrim, a system that ensures real-time, general, efficient, and low-overhead data compaction without compromising attack investigation and anomaly-based intrusion detection. It efficiently reduces log sizes without impacting user experiences, achieving real-time compaction and adaptable deployment on different operating systems. AudiTrim employs two strategies: 1) Data Reduction: By analyzing the types of duplicate edges, our data reduction approach not only considers a broader range of scenarios for redundant edges compared to previous methods but also enhances the efficiency of data reduction. 2) Data Compression: By aggregating log information at the server-side and training a compression model, we facilitate a data compression algorithm that ensures ease of querying. Both strategies meet real-time, low-overhead, and general requirements, fulfilling enterprise data storage needs. The final compaction ratio reaches 26×-65×.
- 30 September 2024
- Published by Association for Computing Machinery (ACM)
- p. 263-277
Open AccessAudiTrim: A Real-time, General, Efficient, and Low-overhead Data Compaction System for Intrusion Detection
Recently enterprises and governments face escalating APT attacks, leading to significant economic losses. APT attacks often persist for extended periods, necessitating the storage of extensive audit logs for effective detection. To reduce data storage overhead, enterprises commonly adopt compression strategies. However, efficient compression strategies may introduce additional query overhead. Existing approaches propose data reduction algorithms, but these methods can compromise data integrity, rendering current attack investigation and anomaly-based intrusion detection ineffective. To address these difficulties, we present AudiTrim, a system that ensures real-time, general, efficient, and low-overhead data compaction without compromising attack investigation and anomaly-based intrusion detection. It efficiently reduces log sizes without impacting user experiences, achieving real-time compaction and adaptable deployment on different operating systems. AudiTrim employs two strategies: 1) Data Reduction: By analyzing the types of duplicate edges, our data reduction approach not only considers a broader range of scenarios for redundant edges compared to previous methods but also enhances the efficiency of data reduction. 2) Data Compression: By aggregating log information at the server-side and training a compression model, we facilitate a data compression algorithm that ensures ease of querying. Both strategies meet real-time, low-overhead, and general requirements, fulfilling enterprise data storage needs. The final compaction ratio reaches 26×-65×.
- Open Access
A New Path: Scaling Vision-and-Language Navigation with Synthetic Instructions and Imitation Learning
Recent studies in Vision-and-Language Navigation (VLN) train RL agents to execute natural-language navigation instructions in photorealistic environments, as a step towards robots that can follow human instructions. However, given the scarcity of human instruction data and limited diversity in the training environments, these agents still struggle with complex language grounding and spatial language understanding. Pretraining on large text and image-text datasets from the web has been extensively explored but the improvements are limited. We investigate large-scale augmentation with synthetic instructions. We take 500+ indoor environments captured in densely-sampled 360 degree panoramas, construct navigation trajectories through these panoramas, and generate a visually-grounded instruction for each trajectory using Marky, a high-quality multilingual navigation instruction generator. We also synthesize image observations from novel viewpoints using an image-to-image GAN. The resulting dataset of 4.2M instruction-trajectory pairs is two orders of magnitude larger than existing human-annotated datasets, and contains a wider variety of environments and viewpoints. To efficiently leverage data at this scale, we train a simple transformer agent with imitation learning. On the challenging RxR dataset, our approach outperforms all existing RL agents, improving the state-of-the-art NDTW from 71.1 to 79.1 in seen environments, and from 64.6 to 66.8 in unseen test environments. Our work points to a new path to improving instruction-following agents, emphasizing large-scale imitation learning and the development of synthetic instruction generation capabilities.
Open AccessA New Path: Scaling Vision-and-Language Navigation with Synthetic Instructions and Imitation Learning
Recent studies in Vision-and-Language Navigation (VLN) train RL agents to execute natural-language navigation instructions in photorealistic environments, as a step towards robots that can follow human instructions. However, given the scarcity of human instruction data and limited diversity in the training environments, these agents still struggle with complex language grounding and spatial language understanding. Pretraining on large text and image-text datasets from the web has been extensively explored but the improvements are limited. We investigate large-scale augmentation with synthetic instructions. We take 500+ indoor environments captured in densely-sampled 360 degree panoramas, construct navigation trajectories through these panoramas, and generate a visually-grounded instruction for each trajectory using Marky, a high-quality multilingual navigation instruction generator. We also synthesize image observations from novel viewpoints using an image-to-image GAN. The resulting dataset of 4.2M instruction-trajectory pairs is two orders of magnitude larger than existing human-annotated datasets, and contains a wider variety of environments and viewpoints. To efficiently leverage data at this scale, we train a simple transformer agent with imitation learning. On the challenging RxR dataset, our approach outperforms all existing RL agents, improving the state-of-the-art NDTW from 71.1 to 79.1 in seen environments, and from 64.6 to 66.8 in unseen test environments. Our work points to a new path to improving instruction-following agents, emphasizing large-scale imitation learning and the development of synthetic instruction generation capabilities.
- Top Cited Papers
- 22 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 17 (15566013) , 3972-3987
THREATRACE: Detecting and Tracing Host-Based Threats in Node Level Through Provenance Graph Learning
Host-based threats such as Program Attack, Malware Implantation, and Advanced Persistent Threats (APT), are commonly adopted by modern attackers. Recent studies propose leveraging the rich contextual information in data provenance to detect threats in a host. Data provenance is a directed acyclic graph constructed from system audit data. Nodes in a provenance graph represent system entities (e.g., processes and files ) and edges represent system calls in the direction of information flow. However, previous studies, which extract features of the provenance graph, are not sensitive to the small quantity of threat-related entities and thus result in low performance when hunting stealthy threats. We present THREATRACE, an anomaly-based detector that detects host-based threats at system entity level without prior knowledge of attack patterns. We tailor GraphSAGE, an inductive graph neural network, to learn every benign entity’s role in a provenance graph. THREATRACE is a real-time system, which is scalable of monitoring a long-term running host and capable of detecting host-based intrusion in their early phase. We evaluate THREATRACE on five public datasets. The results show that THREATRACE outperforms seven state-of-the-art host intrusion detection systems.
- 22 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 17 (15566013) , 3972-3987
Top Cited PapersTHREATRACE: Detecting and Tracing Host-Based Threats in Node Level Through Provenance Graph Learning
Host-based threats such as Program Attack, Malware Implantation, and Advanced Persistent Threats (APT), are commonly adopted by modern attackers. Recent studies propose leveraging the rich contextual information in data provenance to detect threats in a host. Data provenance is a directed acyclic graph constructed from system audit data. Nodes in a provenance graph represent system entities (e.g., processes and files ) and edges represent system calls in the direction of information flow. However, previous studies, which extract features of the provenance graph, are not sensitive to the small quantity of threat-related entities and thus result in low performance when hunting stealthy threats. We present THREATRACE, an anomaly-based detector that detects host-based threats at system entity level without prior knowledge of attack patterns. We tailor GraphSAGE, an inductive graph neural network, to learn every benign entity’s role in a provenance graph. THREATRACE is a real-time system, which is scalable of monitoring a long-term running host and capable of detecting host-based intrusion in their early phase. We evaluate THREATRACE on five public datasets. The results show that THREATRACE outperforms seven state-of-the-art host intrusion detection systems.
- Open Access
Coded Caching with Heterogeneous User Profiles
Coded caching utilizes pre-fetching during off-peak hours and multi-casting for delivery in order to balance the traffic load in communication networks. Several works have studied the achievable peak and average rates under different conditions: variable file lengths or popularities, variable cache sizes, decentralized networks, etc. However, very few have considered the possibility of heterogeneous user profiles, despite modern content providers are investing heavily in categorizing users according to their habits and preferences. This paper proposes three coded caching schemes with uncoded pre-fetching for scenarios where end users are grouped into classes with different file demand sets (FDS). One scheme ignores the difference between the classes, another ignores the intersection between them and the third decouples the delivery of files common to all FDS from those unique to a single class. The transmission rates of the three schemes are compared with a lower bound to evaluate their gap to optimality, and with each other to show that each scheme can outperform the other two when certain conditions are met.
Open AccessCoded Caching with Heterogeneous User Profiles
Coded caching utilizes pre-fetching during off-peak hours and multi-casting for delivery in order to balance the traffic load in communication networks. Several works have studied the achievable peak and average rates under different conditions: variable file lengths or popularities, variable cache sizes, decentralized networks, etc. However, very few have considered the possibility of heterogeneous user profiles, despite modern content providers are investing heavily in categorizing users according to their habits and preferences. This paper proposes three coded caching schemes with uncoded pre-fetching for scenarios where end users are grouped into classes with different file demand sets (FDS). One scheme ignores the difference between the classes, another ignores the intersection between them and the third decouples the delivery of files common to all FDS from those unique to a single class. The transmission rates of the three schemes are compared with a lower bound to evaluate their gap to optimality, and with each other to show that each scheme can outperform the other two when certain conditions are met.
- 1 April 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
LpMAC: A MAC Protocol Based on Valid Prediction of the Next Hop Link in Highly Dynamic Network
Highly dynamic network has a growing demand with the popularization and application of high-speed mobile equipments, such as aircraft, unmanned boat and etc. There are many new features for highly dynamic networks over mobile networks, which include high link failure rate, high frequency enter-and-exit the network and frequent changes of the topology. Therefore, the design of access protocols is more stringent. Particularly, the packets would be invalid when waiting to be sent to the next hop node due to the link breaking with high- speed motion of the nodes in highly dynamic network. In order to better adapt to highly dynamic network scenarios, we propose a novel access protocol based on link validity prediction. Firstly, the protocol obtains the valid time of the packet in the buffer by considering the validity of the next hop link. Further, we designed a backoff mechanism based on packet valid time differentiation. We derive the mathematical expressions for the performance metrics. The performance evaluation exhibits our protocol by providing a lower latency and higher success transmission rate.
- 1 April 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
LpMAC: A MAC Protocol Based on Valid Prediction of the Next Hop Link in Highly Dynamic Network
Highly dynamic network has a growing demand with the popularization and application of high-speed mobile equipments, such as aircraft, unmanned boat and etc. There are many new features for highly dynamic networks over mobile networks, which include high link failure rate, high frequency enter-and-exit the network and frequent changes of the topology. Therefore, the design of access protocols is more stringent. Particularly, the packets would be invalid when waiting to be sent to the next hop node due to the link breaking with high- speed motion of the nodes in highly dynamic network. In order to better adapt to highly dynamic network scenarios, we propose a novel access protocol based on link validity prediction. Firstly, the protocol obtains the valid time of the packet in the buffer by considering the validity of the next hop link. Further, we designed a backoff mechanism based on packet valid time differentiation. We derive the mathematical expressions for the performance metrics. The performance evaluation exhibits our protocol by providing a lower latency and higher success transmission rate.
- 7 February 2019
- Published by Springer Nature
- p. 97-108
Mathematical Modeling and Workspace Analysis for Photographic Robot
- 7 February 2019
- Published by Springer Nature
- p. 97-108
Mathematical Modeling and Workspace Analysis for Photographic Robot