XLXin Liu

Xin 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

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xin Liu has a total of 132 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • MalPDT: Backdoor Attack Against Static Malware Detection With Plug-and-Play Dynamic Triggers

    by Dazhi Zhan, Kun Xu, Xin Liu, Zhisong Pan, Shize Guo, Qi Cao

    The Deep Neural Network (DNN) based detection model’s dependency on third-party crowdsourced sources poses a new security threat from backdoor attacks against malware detectors. Attackers attempt to inject hidden backdoors into the target model, allowing it to perform well on clean samples. Once the attacker-defined trigger activates the hidden backdoor, the model predictions for poisoned samples are maliciously altered. Different from existing backdoor attacks either utilize fixed triggers or generate sample-specific triggers, we explore a novel backdoor attack paradigm in malware domain and propose MalPDT, in which backdoor triggers achieve dynamic variability in trigger patterns and retain compatibility across malware samples. We train a generator capable of hiding information to produce dynamically variable encoded byte segments, which are then injected as triggers into the unused regions of PE malware in a functionality-preserving manner. In MalPDT, any combination of a malware sample and a trigger can form a poisoned sample capable of activating the backdoor, enabling plug-and-play capability. We conduct extensive experiments to validate the effectiveness of MalPDT in attacking models with or without defenses.


  • Open Access

    Enhancing traffic flow prediction through multi-view attention mechanism and dilated convolutional networks

    by Wei Li, Hao Wei, Xin Liu, Jialin Liu, Dazhi Zhan, Xiao Han, Wei Tao

    Accurate traffic flow forecasting serves as a cornerstone for intelligent transportation systems, enabling proactive accident prevention and metropolitan mobility optimization. However, existing approaches face fundamental limitations in modeling the spatiotemporal heterogeneity of traffic dynamics, particularly in simultaneously addressing (1) the decaying significance of temporal dependencies across input sequences and prediction horizons, (2) multi-scale spatial interactions spanning local congestion patterns and global functional correlations, and (3) inter-sample temporal variance in evolving traffic states. To address these limitations, this paper proposes MVA-DCNet (Multi-View Attention Dilated Convolutional Network), a novel deep learning architecture incorporating a multidimensional temporal analysis framework that systematically examines temporal influence mechanisms through three complementary perspectives: inter-sample variance, intra-sequence temporal importance, and output sequence temporal propagation. The proposed model systematically addresses temporal data heterogeneity through three innovative mechanisms: variance-aware data augmentation, adaptive temporal attention, and decaying loss weighting. For enhanced spatial correlation modeling, we develop a dilated convolutional architecture with enhanced receptive field coverage and multi-scale spatial pattern recognition capabilities. Empirical validation on two urban traffic datasets demonstrates superior efficacy in capturing complex spatiotemporal evolution patterns, achieving relative reductions of 12.7% and 9.3% in Root Mean Square Error (RMSE) respectively compared with state-of-the-art benchmarks.

  • Balancing State Exploration and Skill Diversity in Unsupervised Skill Discovery

    by Xin Liu, Yaran Chen, Guixing Chen, Haoran Li, Dongbin Zhao

    Unsupervised skill discovery seeks to acquire different useful skills without extrinsic reward via unsupervised reinforcement learning (RL), with the discovered skills efficiently adapting to multiple downstream tasks in various ways. However, recent advanced skill discovery methods struggle to well balance state exploration and skill diversity, particularly when the potential skills are rich and hard to discern. In this article, we propose contrastive dynamic skill discovery (ComSD) which generates diverse and exploratory unsupervised skills through a novel intrinsic incentive, named contrastive dynamic reward. It contains a particle-based exploration reward to make agents access far-reaching states for exploratory skill acquisition, and a novel contrastive diversity reward to promote the discriminability between different skills. Moreover, a novel dynamic weighting mechanism between the above two rewards is proposed to balance state exploration and skill diversity, which further enhances the quality of the discovered skills. Extensive experiments and analysis demonstrate that ComSD can generate diverse behaviors at different exploratory levels for multijoint robots, enabling state-of-the-art adaptation performance on challenging downstream tasks. It can also discover distinguishable and far-reaching exploration skills in the challenging tree-like 2-D maze.


  • Cross-Domain Random Pretraining With Prototypes for Reinforcement Learning

    by Xin Liu, Yaran Chen, Haoran Li, Boyu Li, Dongbin Zhao

    Unsupervised cross-domain reinforcement learning (RL) pretraining shows great potential for challenging continuous visual control but poses a big challenge. In this article, we propose cross-domain random pretraining with prototypes (CRPTpro), a novel, efficient, and effective self-supervised cross-domain RL pretraining framework. CRPTpro decouples data sampling from encoder pretraining, proposing decoupled random collection to easily and quickly generate a qualified cross-domain pretraining dataset. Moreover, a novel prototypical self-supervised algorithm is proposed to pretrain an effective visual encoder that is generic across different domains. Without finetuning, the cross-domain encoder can be implemented for challenging downstream tasks defined in different domains, either seen or unseen. Compared with recent advanced methods, CRPTpro achieves better performance on downstream policy learning without extra training on exploration agents for data collection, greatly reducing the burden of pretraining. We conduct extensive experiments across multiple challenging continuous visual-control domains, including balance control, robot locomotion, and manipulation. CRPTpro significantly outperforms the next best Proto-RL(C) on 11/12 cross-domain downstream tasks with only 54.5% wall-clock pretraining time, exhibiting state-of-the-art pretraining performance with greatly improved pretraining efficiency.


  • Practical clean-label backdoor attack against static malware detection

    by Dazhi Zhan, Kun Xu, Xin Liu, Tong Han, Zhisong Pan, Shize Guo


  • Open Access

    PSP-Mal: Evading Malware Detection via Prioritized Experience-based Reinforcement Learning with Shapley Prior

    by Dazhi Zhan, Wei Bai, Xin Liu, Yue Hu, Lei Zhang, Shize Guo, Zhisong Pan

    With the widespread application of machine learning techniques in malware detection, researchers have proposed various adversarial attack methods to generate adversarial examples (AEs) of malware, thereby evading detection. Previous studies have shown that the reinforcement learning (RL) framework can enable black-box attacks by performing a sequence of function-preserving operations, which produces functional evasive malware samples. However, it is difficult to obtain the useful guidance and feedbacks from the environment for agent training in the black-box scenario, which results in the RL framework being unable to learn the effective evasion policy. In this paper, we propose the Shapley prior and establish a prior-guidance-based RL framework, namely PSP-Mal, to generate AEs against Portable Executable (PE) malware detectors. Our framework improves on existing methods in three aspects: 1) We explore feature effects of the black-box model by computing Shapley values and further propose the Shapley prior to represent the expected impact of operations. 2) A novel prioritized experience utilization mechanism is established regarding the Shapley prior guidance in the RL framework. 3) The actions are expanded into item-content pairs and we use the Thompson sampling to choose effective content, which helps to reduce randomness and ensure repeatability. We compare the attack performance of our framework with other methods, and experimental results demonstrate that our algorithm is more effective. The evasion rates of PSP-Mal against the LightGBM models trained on EMBER and SOREL-20M reach 76.88% and 72.03%, respectively.


  • [Surgical technique of lateral unicompartmental knee arthroplasty and discussion of the maximum correction value in the treatment of knee valgus deformity].

    by Xin Liu, Kai Zheng, Feng Zhu, Yijun Wang, Lianfang Zhang, Weicheng Zhang, Dechun Geng, Jun Zhou, Yaozeng Xu

    To investigate the surgical technique and the short-term effectivenss of lateral unicompartmental knee arthroplasty (LUKA) through lateral approach in the treatment of valgus knee and to calculate the maximum value of the theoretical correction of knee valgus deformity. A retrospective analysis was performed on 16 patients (20 knees) who underwent LUKA and met the selection criteria between April 2021 and July 2022. There were 2 males and 14 females, aged 57-85 years (mean, 71.5 years). The disease duration ranged from 1 to 18 years, with an average of 11.9 years. Knee valgus was staged according to Ranawat classification, there were 6 knees of type Ⅰ, 13 knees of type Ⅱ, and 1 knee of type Ⅲ. All patients were assigned the expected correction value of genu valgus deformity by preoperative planning, including the correction value of lateral approach, intra-articular correction value, and residual knee valgus deformity value. The actual postoperative corrected values of the above indicators were recorded and the theoretical maximum correctable knee valgus deformity values were extrapolated. The operation time, intraoperative blood loss, incision length, hospital stay, hip-knee-ankle angle (HKA), mechanical lateral distal femoral angle (mLDFA), mechanical medial proximal tibia angle (mMPTA), joint line convergence angle (JLCA), posterior tibial slope (PTS), range of motion (ROM), Hospital for Special Surgery (HSS) score, and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) score were also recorded for effectiveness evaluation. The patients' incision length averaged 13.83 cm, operation time averaged 85.8 minutes, intraoperative blood loss averaged 74.9 mL, and hospital stay averaged 6.7 days. None of the patients suffered any significant intraoperative neurological or vascular injuries. All patients were followed up 10-27 months, with a mean of 17.9 months. One patient with bilateral knee valgus deformities had intra-articular infection in the left knee at 1 month after operation and the remaining patients had no complication such as prosthesis loosening, dislocation, and infection. The ROM, HSS score, and WOMAC score of knee joint significantly improved at each time point after operation when compared to those before operation, and the indicators further improved with time after operation, the differences were all significant ( PP0.05). Postoperative evaluation of the knee valgus deformity correction values showed that the actual intra-articular correction values ranged from 0.54° to 10.97°, with a mean of 3.84°. The postoperative residual knee valgus deformity values ranged from 0.42° to 5.30°, with a mean of 3.59°. The actual correction values of lateral approach ranged from 0.21° to 12.73°, with a mean of 4.26°. LUKA through lateral approach for knee valgus deformity can achieve good early effectiveness. Preoperative planning can help surgeons rationally allocate the correction value of knee valgus deformity, provide corresponding treatment strategies, and the maximum theoretical correction value of knee valgus deformity can reach 25°.


  • Distributed One-Pass Online AUC Maximization

    by Xin Liu, Yu Pan, Yu Ding, Zhisong Pan