QLQiqi Liu

Qiqi Liu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Qiqi Liu has a total of 20 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    MGCL: Tor Website Fingerprinting For Reload Scenario

    by Runhan Song, Qiqi Liu, Junnan Yin, Lingjia Meng, Lei Cui, Wei Wang, Zhenquan Ding, Lun Li, Zhiyu Hao

    Website fingerprinting (WF) attacks aim to identify the websites being visited by analyzing their characteristics. Although prior studies have demonstrated that WF can compromise Tor’s anonymity, they largely assume clean browsing environments. In practice, however, users often revisit pages or navigate within the same site, creating reload traffics where cached resources introduce significant traffic variations. These variations pose challenges to traditional WF methods, yet this issue has received limited attention to date.In this paper, we propose MGCL, a novel Multi-Granularity Contrastive Learning approach for Tor website fingerprinting in reload scenarios. MGCL leverages multi-granularity spatiotemporal features to provide a more comprehensive and robust representation of traffic patterns, thereby enabling more accurate traffic characterization. Then, a dual-branch neural network architecture is designed to process them separately to accommodate their unique semantics and eventually fuse these features. Furthermore, MGCL incorporates supervised contrastive learning to tackle the challenge of distinguishing reload traffic from first-time visit traffic, which belong to the same class but exhibit markedly different feature distributions. By refining feature representations in the vector space, MGCL enhances the model’s robustness to such discrepancies. Experimental results show that MGCL outperforms the state-of-the-art, achieving an accuracy of 85.83%, which is over 10% higher than existing methods.

  • Environment‐Sensitive Carbon Dots Derived from Naphthalenediol for Solvent Polarity Indicator and Anti‐Counterfeiting

    by Jing Bai, Yanyan Cheng, Futao He, Qiqi Liu, Shunfei Qiang, Lanxi Zhang, Jiahao Yang, Chiyu Wang, Yuanqing Xu, Wenkai Zhang

    Herein, a novel type of environment‐sensitive carbon dots (ES‐CDs) has been synthesized via ethanothermal reaction using 1,6‐naphthalenediol as a carbon source and sulfuric acid as a catalyst. The morphology and structure of ES‐CDs are characterized by transmission electron microscopy (TEM), Fourier transform infrared spectroscopy (FT‐IR) and X‐ray photoelectron spectra (XPS). ES‐CDs exhibit green and excitation wavelength independent emission in different solvents. Using the optical response of ES‐CDs towards acetone‐water binary mixtures with different composition, the sensitive detection of solvent polarity is realized based on the inner filter effect (IFE). The ink‐jet printed patterns of ES‐CDs show a blue‐to‐green conversion as exposure to external vapor from hydrogen chloride to ammonia, and enable information to hide assisting with a laser direct writing. We believe that the environment‐sensitive carbon dots developed in this study are potential candidates for solvent sensing and fluorescence anti‐counterfeiting.


  • Binary Vulnerability Mining Based on Long Short-Term Memory Network

    by Wenlei Ouyang, Meng Li, Qiqi Liu, Jianchao Wang

    With the development of human society, the number of software and other computer products began to increase sharply, but the increase in the number of software also brought the increase of the application vulnerabilities behind. In recent years, due to the development of cloud computing and the improvement of computing power, deep learning began to be applied in the field of security, showing new competitiveness in binary program vulnerability detection. In this paper, the idea of natural language processing (NLP) is integrated into vulnerability detection. By extracting function blocks, word2vec is used to get the corresponding feature vector, and LSTM is used to train the obtained feature vector. Through experiments, we verified the potential of vulnerability detection based on deep learning in the future.