Liwei Tian

Dr. Liwei Tian


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 May 2024 in Mathematics
Verified
Journal Article
Open access
PDF file
YCYungui Chen
Yungui Chen
LFLi Feng
Li Feng
QZQinglin Zhao
Qinglin Zhao
LTLiwei Tian
Liwei Tian
LYLei Yang
Lei Yang
E-commerce platforms incorporate reputation systems that allow buyers to rate sellers after transactions. However, existing reputation systems face challenges such as privacy leakage, linkability, and multiple rating attacks. The feedback data can inadvertently expose user information privacy because they reveal the buyers’ identities and preferences, which deters a significant number of users from providing their ratings. Moreover, malicious actors can exploit data analysis and machine learning techniques to mine user privacy from the rating data, posing serious threats to user security and trust. This study introduces ARS-Chain, a pioneering and secure blockchain-driven anonymous reputation-sharing framework tailored for e-commerce platforms. The core of ARS-Chain is a dynamic ring addition mechanism with linkable ring signatures (LRS), where the number of LRS rings is dynamically added in alignment with the evolving purchase list, and LRS link tags are constructed with the LRS rings and item identifiers. Further, a consortium blockchain is introduced to store these anonymous ratings on e-commerce platforms. As a result, ARS-Chain ensures full anonymity while achieving cross-platform reputation sharing, making rating records unlinkable, and effectively countering multiple rating attacks. The experimental results confirm that ARS-Chain significantly enhances user information privacy protection while maintaining system performance, having an important impact on the construction of trust mechanisms for e-commerce platforms.

30 June 2025 in Electronics
Verified
Journal Article
Open access
PDF file
ZXZhiyu Xu
Zhiyu Xu
FLfirst name Lu
first name Lu
YDYong Ding
Yong Ding
LTLiwei Tian
Liwei Tian
SLSuping Liu
Suping Liu
Temporal action localization aims to identify the boundaries of the action of interest in a video. Most existing methods take a two-stage approach: first, identify a set of action proposals; then, based on this set, determine the accurate temporal locations of the action of interest. However, the diversely distributed semantics of a video over time have not been well considered, which could compromise the localization performance, especially for ubiquitous short actions or events (e.g., a fall in healthcare and a traffic violation in surveillance). To address this problem, we propose a novel deep learning architecture, namely an adaptive template-guided self-attention network, to characterize the proposals adaptively with their relevant frames. An input video is segmented into temporal frames, within which the spatio-temporal patterns are formulated by a global–Local Transformer-based encoder. Each frame is associated with a number of proposals of different lengths as their starting frame. Learnable templates for proposals of different lengths are introduced, and each template guides the sampling for proposals with a specific length. It formulates the probabilities for a proposal to form the representation of certain spatio-temporal patterns from its relevant temporal frames. Therefore, the semantics of a proposal can be formulated in an adaptive manner, and a feature map of all proposals can be appropriately characterized. To estimate the IoU of these proposals with ground truth actions, a two-level scheme is introduced. A shortcut connection is also utilized to refine the predictions by using the convolutions of the feature map from coarse to fine. Comprehensive experiments on two benchmark datasets demonstrate the state-of-the-art performance of our proposed method: 32.6% mAP@IoU 0.7 on THUMOS-14 and 9.35% mAP@IoU 0.95 on ActivityNet-1.3.

1 September 2025 in Chinese Journal of Electronics
Verified
Journal Article
Closed access
Ding YongDing Yong
Ding Yong
HLHaifeng Lyu
Haifeng Lyu
YCYanan Cao
Yanan Cao
WSWenxiang Song
Wenxiang Song
LTLiwei Tian
Liwei Tian
SLSuping Liu
Suping Liu
Large language models (LLMs) have transformed natural language processing tasks but remain prone to generating confabulations-plausible yet incorrect or nonsensical outputs. Detecting these confabulations is critical for ensuring the reliability of LLM-driven applications. In this paper, we propose similar cluster frequency entropy (SCFE), a novel entropy-based uncertainty estimator for identifying confabulations in LLM outputs. SCFE works by clustering multiple model-generated responses using cosine similarity to capture similar textual instances, followed by estimating the categorical distribution across these clusters. The uncertainty is quantified through the entropy of this distribution, which reflects the spread of the cluster assignment frequency. High entropy indicates a diffuse probability distribution across clusters, signaling uncertainty, while low entropy suggests a concentrated distribution, indicative of higher confidence. Unlike traditional token-likelihood-based measures, SCFE evaluates uncertainty at a textual-level, capturing meaningful variations across responses. Experimental evaluations on diverse LLMs and benchmark datasets demonstrate that SCFE surpasses existing uncertainty estimation techniques in detecting confabulations, offering a robust tool for enhancing the reliability of LLM outputs in practical applications.

15 November 2025 in Symmetry
Verified
Journal Article
Open access
PDF file
BQBo Qu
Bo Qu
SZSiMin Zheng
SiMin Zheng
JZJunming Zeng
Junming Zeng
LTLiwei Tian
Liwei Tian
Network attacks are becoming increasingly diverse and sophisticated, resulting in complex cybersecurity challenges, which can be fundamentally viewed as a disruption of the symmetry or balanced state in normal network behavior. To address these challenges, graph representation learning methods have gained prominence in network anomaly detection. These methods effectively represent complex network traffic data as graphs and capture data relationships. By integrating deep learning, graph neural networks, and other techniques, graph representation learning enhances the accuracy and efficiency of network anomaly detection in complex network environments. This paper proposes a novel network anomaly detection model based on graph representation learning called ETG-EESAGE. The model constructs an event key time subgraph (ETG) to group similar data and enhance structural features. Then, it introduces an edge enhancement sampling aggregation algorithm (EESAGE) to capture node relations and differentiate edge information accurately. The model generates richer node feature representations during aggregation and detects abnormal nodes using a threshold. Experimental evaluations on the CIC-IDS2017 dataset demonstrate the strong performance of the proposed model across multiple daily subsets. Under optimal configuration settings, ETG-EESAGE achieves an average accuracy of 95.5%, precision of 97.9%, recall of 97.3%, and F1-score of 97.7%, outperforming other baseline algorithms. The model also exhibits strong interpretability and applicability in real-world network anomaly detection scenarios.

21 April 2025 in Electronics
Verified
Journal Article
Open access
PDF file
SWSheng Wen
Sheng Wen
LXLiam Xu
Liam Xu
LTLiwei Tian
Liwei Tian
SLSuping Liu
Suping Liu
Ding YongDing Yong
Ding Yong
The Trusted Execution Environment (TEE) is crucial for safeguarding the ecosystem of embedded systems. It uses isolation to minimize the TCB (Trusted Computing Base) and protect sensitive software. It is vital because devices handle vast, potentially sensitive data. Leveraging ARM TrustZone, widely used in mobile and IoT for TEEs, it ensures hardware protection via security extensions, though needing firmware and software stack support. Despite the reputation of TEEs for high security, TrustZone-aided ones have vulnerabilities. Fuzzing, as a practical bug-finding technique, has seen limited research in the context of TEE. The unique software architecture of TrustZone-assisted TEE complicates the direct application of traditional fuzzing methods. Moreover, simplistic approaches, such as feeding random input values into TEE through the API functions of the rich operating system, fail to uncover deeper, latent bugs within the TEE code. In this paper, we present a fuzzing strategy for TrustZone-assisted TEE that utilizes inferred dependencies between Trusted Kernel system calls to uncover deep-seated TEE bugs. We implemented our approach on OP-TEE, where it successfully identified 17 crashes, including one previously undetected kernel bug.

16 September 2025 in Entropy
Verified
Journal Article
Open access
PDF file
HLHaihua Lu
Haihua Lu
Liang YuLiang Yu
Liang Yu
YHYantao He
Yantao He
LTLiwei Tian
Liwei Tian
Information retrieval serves as a critical methodology for accurately and efficiently obtaining the required information from massive amounts of data. In this paper, we propose an information retrieval framework (SE-MSLC) that utilizes information theory to improve the retrieval effectiveness of inverted index retrieval, thus achieving higher-quality retrieval results in intelligent vertical domain search engines. First, we propose a semantic entropy-driven keyword importance analysis method (SE-KIA) in the query understanding module. This method combines search query logs, the corpus of the search engine, and the theory of semantic entropy, enabling the search engine to dynamically adjust the weights of query keywords, thereby improving its ability to recognize user intent. Then, we propose a hybrid recall strategy that combines a multi-stage strategy and a logical combination strategy (HRS-MSLC) in the recall module. It separately recalls the keywords obtained from the multi-granularity word segmentation of the query in the form of multi-queue recall and simultaneously considers the “AND” and “OR” logical relationships between the keywords. By systematically managing retrieval uncertainty and giving priority to the keywords with high information content, it achieves the best balance between the quantity of the retrieval results and the relevance of the retrieval results to the query. Finally, we experimentally evaluate our methods using the Hit Rate@K and case analysis. Our results demonstrate that the proposed method improves the Hit Rate@1 by 7.3% and the Hit Rate@3 by 6.6% while effectively solving the bad cases in our vertical domain search engine.