AHCI RESEARCH GROUP
Publications
Papers published in international journals,
proceedings of conferences, workshops and books.
OUR RESEARCH
Scientific Publications
How to
Here you can find the complete list of our publications.
You can use the tag cloud to select only the papers dealing with specific research topics.
You can expand the Abstract, Links and BibTex record for each paper.
You can use the tag cloud to select only the papers dealing with specific research topics.
You can expand the Abstract, Links and BibTex record for each paper.
2025
Yang, G.; Sun, Z.; Wang, Y.
ShellBox: Adversarially Enhanced LLM-Interactive Honeypot Framework Journal Article
In: IEEE Access, vol. 13, pp. 143618–143630, 2025, ISSN: 21693536 (ISSN), (Publisher: Institute of Electrical and Electronics Engineers Inc.).
Abstract | Links | BibTeX | Tags: Complex networks, Dynamic error, Dynamic error simulation, Dynamics, Error simulation, Errors, honeypot, Honeypots, Interaction history, Interaction history pruning algorithm, Language Model, Large language model, Multi-turn, Network attack, Network attack and multi-turn, Network Security, Pruning algorithms, Systems analysis, Virtual Reality
@article{yang_shellbox_2025,
title = {ShellBox: Adversarially Enhanced LLM-Interactive Honeypot Framework},
author = {G. Yang and Z. Sun and Y. Wang},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105013277136&doi=10.1109%2FACCESS.2025.3598779&partnerID=40&md5=f776206ec2788fc77e3766f209bc82f1},
doi = {10.1109/ACCESS.2025.3598779},
issn = {21693536 (ISSN)},
year = {2025},
date = {2025-01-01},
journal = {IEEE Access},
volume = {13},
pages = {143618–143630},
abstract = {Honeypot technology is an active defence strategy designed to mitigate the asymmetry inherent in network attacks and defence dynamics. In recent years, honeypot systems powered by large language models (LLMs) have become a focal point of research owing to their ability to simulate complex network environments and generate highly deceptive virtual assets. However, response inconsistency in multi-turn dialogues and prompt injection vulnerabilities inherent to LLMs significantly reduce the deceptive capability of honeypots. This study first defines the threat model, and then introduces a relevance-based interaction history pruning algorithm and dynamic error simulation strategy to mitigate these challenges. Considering practical issues such as response latency and network instability, our experiments were conducted using multiple locally deployed open-source LLMs. The experimental results demonstrated that the proposed dynamic error simulation mechanism achieved a maximum accuracy of 81.63% for the DeepSeek-R1 model. Furthermore, applying the interaction history pruning algorithm improved the turn-level coherence score (TCS) by 34.5% compared with the baseline. Finally, this paper outlines potential future research directions for LLM-based honeypot technologies in active multi-turn mechanisms. © 2025 Elsevier B.V., All rights reserved.},
note = {Publisher: Institute of Electrical and Electronics Engineers Inc.},
keywords = {Complex networks, Dynamic error, Dynamic error simulation, Dynamics, Error simulation, Errors, honeypot, Honeypots, Interaction history, Interaction history pruning algorithm, Language Model, Large language model, Multi-turn, Network attack, Network attack and multi-turn, Network Security, Pruning algorithms, Systems analysis, Virtual Reality},
pubstate = {published},
tppubtype = {article}
}
Honeypot technology is an active defence strategy designed to mitigate the asymmetry inherent in network attacks and defence dynamics. In recent years, honeypot systems powered by large language models (LLMs) have become a focal point of research owing to their ability to simulate complex network environments and generate highly deceptive virtual assets. However, response inconsistency in multi-turn dialogues and prompt injection vulnerabilities inherent to LLMs significantly reduce the deceptive capability of honeypots. This study first defines the threat model, and then introduces a relevance-based interaction history pruning algorithm and dynamic error simulation strategy to mitigate these challenges. Considering practical issues such as response latency and network instability, our experiments were conducted using multiple locally deployed open-source LLMs. The experimental results demonstrated that the proposed dynamic error simulation mechanism achieved a maximum accuracy of 81.63% for the DeepSeek-R1 model. Furthermore, applying the interaction history pruning algorithm improved the turn-level coherence score (TCS) by 34.5% compared with the baseline. Finally, this paper outlines potential future research directions for LLM-based honeypot technologies in active multi-turn mechanisms. © 2025 Elsevier B.V., All rights reserved.
Liu, Y.; Li, Z.
Research on the Design System of Bamboo-Woven Products Based on Traditional Bamboo-Weaving Craft VR Experience Journal Article
In: Forest Products Journal, vol. 75, no. 3, pp. 238–250, 2025, ISSN: 00157473 (ISSN), (Publisher: Forest Products Society).
Abstract | Links | BibTeX | Tags: 3D modeling, Artificial intelligence, Bamboo, Design models, Design systems, evaluation, Experience design, Function evaluation, Human computer interaction, Learn+, Low-costs, Novel techniques, Product design, Product experience, Products, Reliability analysis, Systems analysis, Systems Engineering, Techniques, user experience, User interfaces, Virtual Reality, Virtual reality experiences, Weaving, Weaving technique, Woven products
@article{liu_research_2025,
title = {Research on the Design System of Bamboo-Woven Products Based on Traditional Bamboo-Weaving Craft VR Experience},
author = {Y. Liu and Z. Li},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105013654324&doi=10.13073%2FFPJ-D-25-00005&partnerID=40&md5=d03d78771ee8194ffc8450d259b5f129},
doi = {10.13073/FPJ-D-25-00005},
issn = {00157473 (ISSN)},
year = {2025},
date = {2025-01-01},
journal = {Forest Products Journal},
volume = {75},
number = {3},
pages = {238–250},
abstract = {Virtual reality (VR) is a simulated experience capable of replicating or creating an entirely new environment. Through VR experience, designers can learn bamboo-weaving techniques at a low cost and showcase their design models of bamboo-woven products virtually, allowing these products to be put into production after experience and evaluation. This study introduces novel techniques to transform and innovate traditional bamboo-woven products to establish a comprehensive VR-based product experience design system. This system follows a pioneering pathway, including the following steps: VR weaving skill experience, generative artificial intelligence design (AIGC)–driven bamboo design creativity, 3D modeling technology support, and VR product evaluation. Moreover, the framework conducts user experience research from three dimensions: visual design, system function design, and human–computer interaction design. Usability assessments and statistical analysis were employed before and after the VR experience to assess the system’s reliability. The findings indicate that designers and users can remarkably use and evaluate the new system, offering a practical technical pathway for the modern design exploration of traditional bamboo products. © 2025 Elsevier B.V., All rights reserved.},
note = {Publisher: Forest Products Society},
keywords = {3D modeling, Artificial intelligence, Bamboo, Design models, Design systems, evaluation, Experience design, Function evaluation, Human computer interaction, Learn+, Low-costs, Novel techniques, Product design, Product experience, Products, Reliability analysis, Systems analysis, Systems Engineering, Techniques, user experience, User interfaces, Virtual Reality, Virtual reality experiences, Weaving, Weaving technique, Woven products},
pubstate = {published},
tppubtype = {article}
}
Virtual reality (VR) is a simulated experience capable of replicating or creating an entirely new environment. Through VR experience, designers can learn bamboo-weaving techniques at a low cost and showcase their design models of bamboo-woven products virtually, allowing these products to be put into production after experience and evaluation. This study introduces novel techniques to transform and innovate traditional bamboo-woven products to establish a comprehensive VR-based product experience design system. This system follows a pioneering pathway, including the following steps: VR weaving skill experience, generative artificial intelligence design (AIGC)–driven bamboo design creativity, 3D modeling technology support, and VR product evaluation. Moreover, the framework conducts user experience research from three dimensions: visual design, system function design, and human–computer interaction design. Usability assessments and statistical analysis were employed before and after the VR experience to assess the system’s reliability. The findings indicate that designers and users can remarkably use and evaluate the new system, offering a practical technical pathway for the modern design exploration of traditional bamboo products. © 2025 Elsevier B.V., All rights reserved.