Rongjunchen Zhang
Rongjunchen Zhang has a total of 36 co-authors. Here are the top co-authors by number of shared publications:
- Cecile Paris
Collaborative Intelligence Future Science Platform, CSIRO Data61, Australia
3 shared publications - Sheng Wen
Swinburne University of Technology, Melbourne, Australia
3 shared publications - Surya Nepal
Data61, CSIRO, Canberra, ACT, Australia
3 shared publications - Yang Xiang
Swinburne University of Technology
3 shared publications - Dong Zhang
HiThink Research, Hangzhou, China
2 shared publications - Haipang Wu
HiThink Research, Hangzhou, China
2 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 9058-9067
Compress to Focus: Efficient Coordinate Compression for Policy Optimization in Multi-Turn GUI Agents
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 9058-9067
Open AccessCompress to Focus: Efficient Coordinate Compression for Policy Optimization in Multi-Turn GUI Agents
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 10787-10796
NEXUS-O: An Omni-Perceptive and -Interactive Model for Language, Audio, and Vision
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 10787-10796
Open AccessNEXUS-O: An Omni-Perceptive and -Interactive Model for Language, Audio, and Vision
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12867-12874
MME-Finance: A Multimodal Finance Benchmark for Expert-level Understanding and Reasoning
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 12867-12874
Open AccessMME-Finance: A Multimodal Finance Benchmark for Expert-level Understanding and Reasoning
- Open Access
- 30 April 2023
- Published by Association for Computing Machinery (ACM)
- p. 1604-1615
Dynalogue: A Transformer-Based Dialogue System with Dynamic Attention
Businesses face a range of cyber risks, both external threats and internal vulnerabilities that continue to evolve over time. As cyber attacks continue to increase in complexity and sophistication, more organisations will experience them. For this reason, it is important that organisations seek timely consultancy from cyber professionals so that they can respond to and recover from cyber attacks as quickly as possible. However, huge surges in cyber attacks have long left cyber professionals short of what is required to cover the security needs. This problem is getting worse when an increasing number of people choose to work from home during the pandemic because this situation usually yields extra communication cost. In this paper, we propose to develop a cybersecurity-oriented dialogue system, called Dynalogue1, which can provide consultancy online as a cyber professional. For the first time, Dynalogue provides a promising solution to mitigate the need for cyber professionals via automatically generating problem-targeted conversions to victims of cyber attacks. In spite of many dialogue systems developed in the past, Dynalogue provides a distinct capability of handling long and complicated sentences that are common in cybersecurity-related conversations. It is challenging to have this capability because limited memory in dialogue systems can be hard to accommodate sufficient key information of long sentences. To overcome this challenge, Dynalogue utilises an attention mechanism that dynamically captures key semantics within a sentence instead of using fix window to cut off the sentence. To evaluate Dynalogue, we collect 67K real-world conversations (0.6M utterances) from Bleeping Computer2, which is one of the most popular cybersecurity consultancy websites in the world. The results suggest that Dynalogue outperforms all the existing dialogue systems with 1% ∼ 9% improvements on all different metrics. We further run Dynalogue on the public dataset WikiHow to validate its compatibility in other domains where conversations are also long and complicated. Dynalogue also outperforms all the other methods with at most 2.4% improvement.
- 30 April 2023
- Published by Association for Computing Machinery (ACM)
- p. 1604-1615
Open AccessDynalogue: A Transformer-Based Dialogue System with Dynamic Attention
Businesses face a range of cyber risks, both external threats and internal vulnerabilities that continue to evolve over time. As cyber attacks continue to increase in complexity and sophistication, more organisations will experience them. For this reason, it is important that organisations seek timely consultancy from cyber professionals so that they can respond to and recover from cyber attacks as quickly as possible. However, huge surges in cyber attacks have long left cyber professionals short of what is required to cover the security needs. This problem is getting worse when an increasing number of people choose to work from home during the pandemic because this situation usually yields extra communication cost. In this paper, we propose to develop a cybersecurity-oriented dialogue system, called Dynalogue1, which can provide consultancy online as a cyber professional. For the first time, Dynalogue provides a promising solution to mitigate the need for cyber professionals via automatically generating problem-targeted conversions to victims of cyber attacks. In spite of many dialogue systems developed in the past, Dynalogue provides a distinct capability of handling long and complicated sentences that are common in cybersecurity-related conversations. It is challenging to have this capability because limited memory in dialogue systems can be hard to accommodate sufficient key information of long sentences. To overcome this challenge, Dynalogue utilises an attention mechanism that dynamically captures key semantics within a sentence instead of using fix window to cut off the sentence. To evaluate Dynalogue, we collect 67K real-world conversations (0.6M utterances) from Bleeping Computer2, which is one of the most popular cybersecurity consultancy websites in the world. The results suggest that Dynalogue outperforms all the existing dialogue systems with 1% ∼ 9% improvements on all different metrics. We further run Dynalogue on the public dataset WikiHow to validate its compatibility in other domains where conversations are also long and complicated. Dynalogue also outperforms all the other methods with at most 2.4% improvement.
- Open Access
- 28 February 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Internet Technology
- Vol. 23 (1) , 1-18
SAM: Multi-turn Response Selection Based on Semantic Awareness Matching
Multi-turn response selection is a key issue in retrieval-based chatbots and has attracted considerable attention in the NLP (Natural Language processing) field. So far, researchers have developed many solutions that can select appropriate responses for multi-turn conversations. However, these works are still suffering from the semantic mismatch problem when responses and context share similar words with different meanings. In this article, we propose a novel chatbot model based on Semantic Awareness Matching, called SAM. SAM can capture both similarity and semantic features in the context by a two-layer matching network. Appropriate responses are selected according to the matching probability made through the aggregation of the two feature types. In the evaluation, we pick 4 widely used datasets and compare SAM’s performance to that of 12 other models. Experiment results show that SAM achieves substantial improvements, with up to 1.5% R 10 @1 on Ubuntu Dialogue Corpus V2, 0.5% R 10 @1 on Douban Conversation Corpus, and 1.3% R 10 @1 on E-commerce Corpus.
- 28 February 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Internet Technology
- Vol. 23 (1) , 1-18
Open AccessSAM: Multi-turn Response Selection Based on Semantic Awareness Matching
Multi-turn response selection is a key issue in retrieval-based chatbots and has attracted considerable attention in the NLP (Natural Language processing) field. So far, researchers have developed many solutions that can select appropriate responses for multi-turn conversations. However, these works are still suffering from the semantic mismatch problem when responses and context share similar words with different meanings. In this article, we propose a novel chatbot model based on Semantic Awareness Matching, called SAM. SAM can capture both similarity and semantic features in the context by a two-layer matching network. Appropriate responses are selected according to the matching probability made through the aggregation of the two feature types. In the evaluation, we pick 4 widely used datasets and compare SAM’s performance to that of 12 other models. Experiment results show that SAM achieves substantial improvements, with up to 1.5% R 10 @1 on Ubuntu Dialogue Corpus V2, 0.5% R 10 @1 on Douban Conversation Corpus, and 1.3% R 10 @1 on E-commerce Corpus.
- Open Access
Catering to Your Concerns: Automatic Generation of Personalised Security-Centric Descriptions for Android Apps
Android users are increasingly concerned with the privacy of their data and security of their devices. To improve the security awareness of users, recent automatic techniques produce security-centric descriptions by performing program analysis. However, the generated text does not always address users' concerns as they are generally too technical to be understood by ordinary users. Moreover, different users have varied linguistic preferences, which do not match the text. Motivated by this challenge, we develop an innovative scheme to help users avoid malware and privacy-breaching apps by generating security descriptions that explain the privacy and security related aspects of an Android app in clear and understandable terms. We implement a prototype system, PERSCRIPTION, to generate personalised security-centric descriptions that automatically learn users' security concerns and linguistic preferences to produce user-oriented descriptions. We evaluate our scheme through experiments and user studies. The results clearly demonstrate the improvement on readability and users' security awareness of PERSCRIPTION's descriptions compared to existing description generators.
Open AccessCatering to Your Concerns: Automatic Generation of Personalised Security-Centric Descriptions for Android Apps
Android users are increasingly concerned with the privacy of their data and security of their devices. To improve the security awareness of users, recent automatic techniques produce security-centric descriptions by performing program analysis. However, the generated text does not always address users' concerns as they are generally too technical to be understood by ordinary users. Moreover, different users have varied linguistic preferences, which do not match the text. Motivated by this challenge, we develop an innovative scheme to help users avoid malware and privacy-breaching apps by generating security descriptions that explain the privacy and security related aspects of an Android app in clear and understandable terms. We implement a prototype system, PERSCRIPTION, to generate personalised security-centric descriptions that automatically learn users' security concerns and linguistic preferences to produce user-oriented descriptions. We evaluate our scheme through experiments and user studies. The results clearly demonstrate the improvement on readability and users' security awareness of PERSCRIPTION's descriptions compared to existing description generators.
- Open Access
- 7 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 153542-153554
Using AI to Attack VA: A Stealthy Spyware Against Voice Assistances in Smart Phones
Voice Assistants (VAs) are increasingly popular for human-computer interaction (HCI) smartphones. To help users automatically conduct various tasks, these tools usually come with high privileges and are able to access sensitive system resources. A comprised VA is a stepping stone for attackers to hack into users’ phones. Prior work has experimentally demonstrated that VAs can be a promising attack point for HCI tools. However, the state-of-the-art approaches require ad-hoc mechanisms to activate VAs that are non-trivial to trigger in practice and are usually limited to specific mobile platforms. To mitigate the limitations faced by the state-of-the-art, we propose a novel attack approach, namely Vaspy, which crafts the users’ “activation voice” by silently listening to users’ phone calls. Once the activation voice is formed, Vaspy can select a suitable occasion to launch an attack. Vaspy embodies a machine learning model that learns suitable attacking times to prevent the attack from being noticed by the user. We implement a proof-of-concept spyware and test it on a range of popular Android phones. The experimental results demonstrate that this approach can silently craft the activation voice of the users and launch attacks. In the wrong hands, a technique like Vaspy can enable automated attacks to HCI tools. By raising awareness, we urge the community and manufacturers to revisit the risks of VAs and subsequently revise the activation logic to be resilient to the style of attacks proposed in this work.
- 7 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 153542-153554
Open AccessUsing AI to Attack VA: A Stealthy Spyware Against Voice Assistances in Smart Phones
Voice Assistants (VAs) are increasingly popular for human-computer interaction (HCI) smartphones. To help users automatically conduct various tasks, these tools usually come with high privileges and are able to access sensitive system resources. A comprised VA is a stepping stone for attackers to hack into users’ phones. Prior work has experimentally demonstrated that VAs can be a promising attack point for HCI tools. However, the state-of-the-art approaches require ad-hoc mechanisms to activate VAs that are non-trivial to trigger in practice and are usually limited to specific mobile platforms. To mitigate the limitations faced by the state-of-the-art, we propose a novel attack approach, namely Vaspy, which crafts the users’ “activation voice” by silently listening to users’ phone calls. Once the activation voice is formed, Vaspy can select a suitable occasion to launch an attack. Vaspy embodies a machine learning model that learns suitable attacking times to prevent the attack from being noticed by the user. We implement a proof-of-concept spyware and test it on a range of popular Android phones. The experimental results demonstrate that this approach can silently craft the activation voice of the users and launch attacks. In the wrong hands, a technique like Vaspy can enable automated attacks to HCI tools. By raising awareness, we urge the community and manufacturers to revisit the risks of VAs and subsequently revise the activation logic to be resilient to the style of attacks proposed in this work.
- 9 September 2019
- Published by Springer Nature
- p. 378-396
Who Activated My Voice Assistant? A Stealthy Attack on Android Phones Without Users’ Awareness
- 9 September 2019
- Published by Springer Nature
- p. 378-396
Who Activated My Voice Assistant? A Stealthy Attack on Android Phones Without Users’ Awareness