Tianhao Peng

Tianhao Peng

Auto-generated profile, based on publicly available information

Tianhao Peng

Beihang University, Beijing, China

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
4 February 2025 in IEEE Transactions on Knowledge and Data Engineering
Verified
Journal Article
Closed access
TPTianhao Peng
Tianhao Peng
HYHaitao Yuan
Haitao Yuan
YZYongqi Zhang
Yongqi Zhang
YLYuchen Li
Yuchen Li
PDPeihong Dai
Peihong Dai
Wenjun WuWenjun Wu
Wenjun Wu
Sequential recommendation systems aim to predict the future behaviors of users based on their historical interactions. Despite the success of neural architectures like Transformer and Graph Neural Networks, these models often struggle with the inherent challenge of sparse data in accurately predicting future user behaviors. To alleviate the data sparsity problem, some methods leverage the contrastive learning to generate contrastive views, assuming the items appear discretely at the same time intervals and focusing on the sequence order. However, these approaches neglect the crucial temporal-aware collaborative patterns hidden within the user-item interactions, leading to a limited variety of contrastive pairs and less informative embeddings. The proposed framework, Temporal-aware graph contrastive learning with theoretical guarantees for sequential Recommendation (TagRec), integrates temporal-aware collaborative patterns with adaptive data augmentation to generate more informative user and item representations. TagRec employs a temporal-aware graph neural network to embed the original graph, then generates augmented graphs through the addition of interactions via latent user interest mining, the dropping of redundant interaction edges, and the perturbation of temporal information. Theoretical guarantees are provided that these augmentations enhance the graph's utility. Extensive experiments on real-world datasets demonstrate the superiority of the proposed approach over the state-of-the-art recommendation methods.

26 April 2024 in ACM Transactions on Information Systems
Verified
Journal Article
Open access
YPYanjun Pu
Yanjun Pu
FLFang Liu
Fang Liu
RSRongye Shi
Rongye Shi
HYHaitao Yuan
Haitao Yuan
RCRuibo Chen
Ruibo Chen
Wenjun WuWenjun Wu
Wenjun Wu
Knowledge tracing models based on deep learning can achieve impressive predictive performance by leveraging attention mechanisms. However, there still exist two challenges in attentive knowledge tracing (AKT): First, the mechanism of classical models of AKT demonstrates relatively low attention when processing exercise sequences with shifting knowledge concepts (KC), making it difficult to capture the comprehensive state of knowledge across sequences. Second, classical models do not consider stochastic behaviors, which negatively affects models of AKT in terms of capturing anomalous knowledge states. This article proposes a model of AKT, called Enhancing Locality for Attentive Knowledge Tracing (ELAKT), that is a variant of the deep KT model. The proposed model leverages the encoder module of the transformer to aggregate knowledge embedding generated by both exercises and responses over all timesteps. In addition, it uses causal convolutions to aggregate and smooth the states of local knowledge. The ELAKT model uses the states of comprehensive KCs to introduce a prediction correction module to forecast the future responses of students to deal with noise caused by stochastic behaviors. The results of experiments demonstrated that the ELAKT model consistently outperforms state-of-the-art baseline KT models.

20 July 2025
Verified
Conference Paper
Open access
YLYuchen Li
Yuchen Li
HXHaoyi Xiong
Haoyi Xiong
YZYongqi Zhang
Yongqi Zhang
JBJiang Bian
Jiang Bian
TPTianhao Peng
Tianhao Peng
DYDawei Yin
Dawei Yin

21 March 2025 in Scientific Reports
Verified
Journal Article
Open access
PDF file
TPTianhao Peng
Tianhao Peng
QYQiang Yue
Qiang Yue
Yu LiangYu Liang
Yu Liang
JRJian Ren
Jian Ren
JLJie Luo
Jie Luo
Wenjun WuWenjun Wu
Wenjun Wu
Collaborative learning is a prevalent learning method, and modeling and predicting student performance in such paradigms is an important task. Most current methods analyze this complex task solely based on the frequency of student activities, overlooking the rich spatial and temporal features present in these activities, as well as the diverse textual content provided by various learning artifacts. To address these challenges, we choose a software engineering course as the study subject, where students are required to team up and complete a software project together. In this paper, we propose a novel Global-local Optimized grAph Transformer framework for collaborative learning, termed GOAT. Specifically, we first construct the dynamic knowledge concept-enhanced interaction graphs with nodes representing both students and relevant software engineering concepts, and edges illustrating interactions. Additionally, we incorporate spatial-aware and temporal-aware modules to capture the respective information, enabling the modeling of dynamic interactions within and across learning teams over time. A global-local optimization module is introduced to model intricate relationships within and between teams, highlighting commonalities and differences among team members. Our framework is backed by theoretical analysis and validated through extensive experiments on real-world datasets, which demonstrate its superiority over existing methods.

20 October 2025 in IEEE Transactions on Knowledge and Data Engineering
Verified
Journal Article
Closed access
YLYuchen Li
Yuchen Li
ZLZhonghao Lyu
Zhonghao Lyu
YZYongqi Zhang
Yongqi Zhang
HZHao Zhang
Hao Zhang
TPTianhao Peng
Tianhao Peng
DYDawei Yin
Dawei Yin
Learning-to-Rank (LTR) models built on Transformers have been widely adopted to achieve commendable performance in web search. However, these models predominantly emphasize relevance, often overlooking broader aspects of user satisfaction such as quality, authority, and recency, which collectively enhance the overall user experience. Addressing these multifaceted elements is essential for developing more effective and user-centric search engines. Nevertheless, training such comprehensive models remains challenging due to the scarcity of annotated query-webpage pairs relative to the vast number of webpages available online and the billions of daily search queries. Concurrently, industry research communities have released numerous open-source LTR datasets with well-annotated samples, though these datasets feature diverse designs of LTR features and labels across heterogeneous domains. Inspired by recent advancements in pre-training transformers for enhanced performance, this work explores the pre-training of LTR models using both labeled and unlabeled samples. Specifically, we leverage well-annotated samples from heterogeneous open-source LTR datasets to bolster the pre-training process and integrate multifaceted satisfaction features during the fine-tuning stage. In this paper, we propose S$^{3}$3PRank—Satisfaction-oriented Learning to Rank with Semi-supervised Pre-training. Specifically, S$^{3}$PRank employs a three-step approach: (1) it exploits unlabeled/labeled data from the search engine to pre-train a self-attentive encoder via semi-supervised learning; (2) it incorporates multiple open-source heterogeneous LTR datasets to enhance the pre-training of the relevance tower through shared parameters in cross-domain learning; (3) it integrates a satisfaction tower with the pre-trained relevance tower to form a deep two-tower aggregation structure, and fine-tunes the combination of pre-trained self-attentive encoder and the two-tower structure using search engine data with various learning strategies. To demonstrate the effectiveness of our proposed approach, we conduct extensive offline and online evaluations using real-world web traffic from Baidu Search. The comparisons against numbers of advanced baselines confirmed the advantages of S$^{3}$PRank in producing high-performance ranking models for web-scale search.

27 February 2026 in IEEE Journal of Biomedical and Health Informatics
Verified
Journal Article
Closed access
RCRongqi Cao
Rongqi Cao
JHJian He
Jian He
Yu LiangYu Liang
Yu Liang
XHXiyuan Hu
Xiyuan Hu
TPTianhao Peng
Tianhao Peng
Shahid MumtazShahid Mumtaz
Shahid Mumtaz
Electroencephalogram (EEG)-based emotion recognition systems face a persistent challenge in maintaining robust performance across subjects (generalization) and within subjects (personalization). Existing models for cross-subject recognition generally struggle to adapt to individual-specific neural signatures, while models with optimized within-subject performance typically require a large amount of personalized data. To address these limitations, this study proposes an EEG-based emotion recognition framework, CLDAE, that integrates a contrastive learning strategy and a dual-attention feature extraction mechanism. The CLDAE framework includes two stages: contrastive learning pre-training and emotion recognition fine-tuning. During the pre-training stage, a data augmentation method that combines EEG signals from different subjects is used to generate new training samples. Moreover, to extract discriminative features from the augmented data, the dual-attention encoder combines temporal and channel attention mechanisms. After pre-training, the CLDAE is fine-tuned for final recognition tasks. The proposed CLDAE is verified by experiments on two public datasets (DEAP and SEED-IV) and a private dataset (MAN). The experimental results demonstrate that the CLDAE achieves competitive performance in both within-subject and cross-subject emotion recognition, with 95.12% and 75.29% accuracy on the MAN dataset, respectively; thus, outperforming the baseline methods. These results validate the effectiveness of the proposed framework in both within-subject and cross-subject emotion recognition.