Shujie Li
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
Shujie Li has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Edith Cheuk Han Ngai
The University of Hong Kong, Hong Kong, Hong Kong
3 shared publications - Guanghu Yuan
Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
3 shared publications - Jinfeng Xu
The University of Hong Kong, Hong Kong, Hong Kong
3 shared publications - Chengming Li
Shenzhen MSU-BIT University, Shenzhen, China
2 shared publications - Feng Liang
Shenzhen MSU-BIT University, Shenzhen, China
2 shared publications - Jian Chen
The University of Hong Kong
2 shared publications
Latest Publications
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 2753-2764
Learning Semantic Co-occurrence and Personalized Focused Representations for Multimodal Sequential Recommendation
Multimodal sequential recommendation (MSR) leverages textual and visual content of items to capture the evolution of user interests over time. Prior efforts have largely concentrated on multimodal fusion strategies or downstream sequential modeling, while the question of how to obtain item representations genuinely aligned with the recommendation objective has received far less attention. Off-the-shelf encoders pretrained on generic corpora produce representations that emphasize broad semantic similarity rather than the specific cues that drive user choice in a recommendation scenario. In this work, we propose ScopeRec, a Semantic co-occurrence and personalized focus framework for multimodal sequential Recommendation, which reshapes multimodal representations from two complementary views. From the item view, items co-interacted within a behavior sequence carry rich sequence-level semantic associations, which we term semantic co-occurrence. ScopeRec mines such multi-level co-occurrence patterns directly from historical sequences and injects them back into the item representations, yielding embeddings that better reflect users' historical semantic preferences. From the user view, user attention typically concentrates on a few keywords in text or salient regions in images, which we term semantic focus. To exploit this property, ScopeRec employs a hierarchical mixture-of-experts (MoE) architecture in which different experts independently capture fine-grained interest from distinct semantic perspectives. A specialized fusion module then adaptively combines representations across modalities and semantic levels in a personalized manner. Experiments on three public MSR benchmarks show that ScopeRec consistently surpasses state-of-the-art baselines, with relative gains of 13.8% to 28.7% on NDCG and Hit Ratio.
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 2753-2764
Open AccessLearning Semantic Co-occurrence and Personalized Focused Representations for Multimodal Sequential Recommendation
Multimodal sequential recommendation (MSR) leverages textual and visual content of items to capture the evolution of user interests over time. Prior efforts have largely concentrated on multimodal fusion strategies or downstream sequential modeling, while the question of how to obtain item representations genuinely aligned with the recommendation objective has received far less attention. Off-the-shelf encoders pretrained on generic corpora produce representations that emphasize broad semantic similarity rather than the specific cues that drive user choice in a recommendation scenario. In this work, we propose ScopeRec, a Semantic co-occurrence and personalized focus framework for multimodal sequential Recommendation, which reshapes multimodal representations from two complementary views. From the item view, items co-interacted within a behavior sequence carry rich sequence-level semantic associations, which we term semantic co-occurrence. ScopeRec mines such multi-level co-occurrence patterns directly from historical sequences and injects them back into the item representations, yielding embeddings that better reflect users' historical semantic preferences. From the user view, user attention typically concentrates on a few keywords in text or salient regions in images, which we term semantic focus. To exploit this property, ScopeRec employs a hierarchical mixture-of-experts (MoE) architecture in which different experts independently capture fine-grained interest from distinct semantic perspectives. A specialized fusion module then adaptively combines representations across modalities and semantic levels in a personalized manner. Experiments on three public MSR benchmarks show that ScopeRec consistently surpasses state-of-the-art baselines, with relative gains of 13.8% to 28.7% on NDCG and Hit Ratio.
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 7991-7996
Neural Tree Collaborative Filtering: Rethinking Graph Collaborative Filtering as Tree Collaborative Filtering with Curvature-Aware Propagation Depth
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 7991-7996
Open AccessNeural Tree Collaborative Filtering: Rethinking Graph Collaborative Filtering as Tree Collaborative Filtering with Curvature-Aware Propagation Depth
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 5898-5909
When Is Shallow Enough? Adaptive Split Federated Learning with Client-Specific Sufficiency Estimation
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 5898-5909
Open AccessWhen Is Shallow Enough? Adaptive Split Federated Learning with Client-Specific Sufficiency Estimation
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (6) , 12491-12500
FedSM: Semantic-Guided Feature Mixup for Bias Reduction in Federated Learning With Long-Tail Data
Federated Learning (FL) has emerged as a promising paradigm for decentralized machine learning, where a central server coordinates distributed clients to collaboratively train a global model without direct access to raw data. Despite its advantages, heterogeneous and long-tail data distributions across clients remain a major bottleneck, particularly in IoT scenarios with diverse devices and sensing modalities. To address these challenges, we propose FedSM, a novel framework that integrates multimodal semantic knowledge with balanced pseudo features to enhance global model optimization. Unlike conventional approaches that rely on single-modal information, FedSM leverages CLIP’s cross-modal representations and open-vocabulary priors to guide semantic-aware data augmentation. A probabilistic selection mechanism further refines local features by mixing them with global prototypes, ensuring pseudo features are semantically reliable and reducing bias caused by skewed client distributions. Almost all computations are performed locally at the client side, thereby alleviating server overhead and improving scalability in resource-constrained IoT environments. Extensive experiments on long-tail benchmarks including CIFAR-10-LT, CIFAR-100-LT, and ImageNet-LT demonstrate the superiority of FedSM over state-of-the-art baselines, highlighting its potential for robust communication-efficient FL in IoT networks.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (6) , 12491-12500
FedSM: Semantic-Guided Feature Mixup for Bias Reduction in Federated Learning With Long-Tail Data
Federated Learning (FL) has emerged as a promising paradigm for decentralized machine learning, where a central server coordinates distributed clients to collaboratively train a global model without direct access to raw data. Despite its advantages, heterogeneous and long-tail data distributions across clients remain a major bottleneck, particularly in IoT scenarios with diverse devices and sensing modalities. To address these challenges, we propose FedSM, a novel framework that integrates multimodal semantic knowledge with balanced pseudo features to enhance global model optimization. Unlike conventional approaches that rely on single-modal information, FedSM leverages CLIP’s cross-modal representations and open-vocabulary priors to guide semantic-aware data augmentation. A probabilistic selection mechanism further refines local features by mixing them with global prototypes, ensuring pseudo features are semantically reliable and reducing bias caused by skewed client distributions. Almost all computations are performed locally at the client side, thereby alleviating server overhead and improving scalability in resource-constrained IoT environments. Extensive experiments on long-tail benchmarks including CIFAR-10-LT, CIFAR-100-LT, and ImageNet-LT demonstrate the superiority of FedSM over state-of-the-art baselines, highlighting its potential for robust communication-efficient FL in IoT networks.
- Open Access
- 21 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 3063-3072
MMLRec: A Unified Multi-Task and Multi-Scenario Learning Benchmark for Recommendation
In recent years, there has been a trend in the field of recommender systems towards multi-task modeling and multi-scenario modeling. The aim is to enhance the performance of various tasks and scenarios by jointly training on multiple tasks or scenarios to learn common patterns and features. Joint modeling of tasks and scenarios has also received widespread attention recently. However, despite the rich proposals of methods for Multi-Task Learning (MTL), Multi-Scenario Learning (MSL), and Multi-Task-Multi-Scenario Learning (MTMSL) in recent years, there still lacks a comprehensive benchmark to evaluate these methods. Previous studies often employed different datasets, data processing techniques, data partitioning strategies, and hyperparameter settings, making replication of existing research and fair comparison of experimental results challenging. To address this challenge, we introduce MMLRec, the first unified comprehensive benchmark for evaluating MTL, MSL and MTMSL, featuring consistent dataset processing and identical parameter settings. This benchmark implements a range of MTL, MSL, and MTMSL algorithms, and evaluates them on multiple commonly used recommender systems datasets. Through fair comparative experiments, we find that some structurally simplistic recommendation algorithms are underestimated, as they can achieve comparable results to more complex algorithms while maintaining lower complexity. Furthermore, our experimental analysis indicates that more complex methods exhibit better robustness when there are significant differences between tasks or scenarios. By providing a unified framework (MMLRec), our goal is to promote rapid evaluation and inspire innovative research in this continuously evolving field. We hope that our open-source benchmark can facilitate swift, equitable evaluations, while also fostering further breakthrough research in the domains of MTL, MSL, and MTMSL.
- 21 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 3063-3072
Open AccessMMLRec: A Unified Multi-Task and Multi-Scenario Learning Benchmark for Recommendation
In recent years, there has been a trend in the field of recommender systems towards multi-task modeling and multi-scenario modeling. The aim is to enhance the performance of various tasks and scenarios by jointly training on multiple tasks or scenarios to learn common patterns and features. Joint modeling of tasks and scenarios has also received widespread attention recently. However, despite the rich proposals of methods for Multi-Task Learning (MTL), Multi-Scenario Learning (MSL), and Multi-Task-Multi-Scenario Learning (MTMSL) in recent years, there still lacks a comprehensive benchmark to evaluate these methods. Previous studies often employed different datasets, data processing techniques, data partitioning strategies, and hyperparameter settings, making replication of existing research and fair comparison of experimental results challenging. To address this challenge, we introduce MMLRec, the first unified comprehensive benchmark for evaluating MTL, MSL and MTMSL, featuring consistent dataset processing and identical parameter settings. This benchmark implements a range of MTL, MSL, and MTMSL algorithms, and evaluates them on multiple commonly used recommender systems datasets. Through fair comparative experiments, we find that some structurally simplistic recommendation algorithms are underestimated, as they can achieve comparable results to more complex algorithms while maintaining lower complexity. Furthermore, our experimental analysis indicates that more complex methods exhibit better robustness when there are significant differences between tasks or scenarios. By providing a unified framework (MMLRec), our goal is to promote rapid evaluation and inspire innovative research in this continuously evolving field. We hope that our open-source benchmark can facilitate swift, equitable evaluations, while also fostering further breakthrough research in the domains of MTL, MSL, and MTMSL.
- Open Access
- 26 April 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 42 (4) , 1-28
Improving Semi-Supervised Text Classification with Dual Meta-Learning
The goal of semi-supervised text classification (SSTC) is to train a model by exploring both a small number of labeled data and a large number of unlabeled data, such that the learned semi-supervised classifier performs better than the supervised classifier trained on solely the labeled samples. Pseudo-labeling is one of the most widely used SSTC techniques, which trains a teacher classifier with a small number of labeled examples to predict pseudo labels for the unlabeled data. The generated pseudo-labeled examples are then utilized to train a student classifier, such that the learned student classifier can outperform the teacher classifier. Nevertheless, the predicted pseudo labels may be inaccurate, making the performance of the student classifier degraded. The student classifier may perform even worse than the teacher classifier. To alleviate this issue, in this paper, we introduce a dual meta-learning ( DML ) technique for semi-supervised text classification, which improves the teacher and student classifiers simultaneously in an iterative manner. Specifically, we propose a meta-noise correction method to improve the student classifier by proposing a Noise Transition Matrix (NTM) with meta-learning to rectify the noisy pseudo labels. In addition, we devise a meta pseudo supervision method to improve the teacher classifier. Concretely, we exploit the feedback performance from the student classifier to further guide the teacher classifier to produce more accurate pseudo labels for the unlabeled data. In this way, both teacher and student classifiers can co-evolve in the iterative training process. Extensive experiments on four benchmark datasets highlight the effectiveness of our DML method against existing state-of-the-art methods for semi-supervised text classification. We release our code and data of this paper publicly at https://github.com/GRIT621/DML.
- 26 April 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 42 (4) , 1-28
Open AccessImproving Semi-Supervised Text Classification with Dual Meta-Learning
The goal of semi-supervised text classification (SSTC) is to train a model by exploring both a small number of labeled data and a large number of unlabeled data, such that the learned semi-supervised classifier performs better than the supervised classifier trained on solely the labeled samples. Pseudo-labeling is one of the most widely used SSTC techniques, which trains a teacher classifier with a small number of labeled examples to predict pseudo labels for the unlabeled data. The generated pseudo-labeled examples are then utilized to train a student classifier, such that the learned student classifier can outperform the teacher classifier. Nevertheless, the predicted pseudo labels may be inaccurate, making the performance of the student classifier degraded. The student classifier may perform even worse than the teacher classifier. To alleviate this issue, in this paper, we introduce a dual meta-learning ( DML ) technique for semi-supervised text classification, which improves the teacher and student classifiers simultaneously in an iterative manner. Specifically, we propose a meta-noise correction method to improve the student classifier by proposing a Noise Transition Matrix (NTM) with meta-learning to rectify the noisy pseudo labels. In addition, we devise a meta pseudo supervision method to improve the teacher classifier. Concretely, we exploit the feedback performance from the student classifier to further guide the teacher classifier to produce more accurate pseudo labels for the unlabeled data. In this way, both teacher and student classifiers can co-evolve in the iterative training process. Extensive experiments on four benchmark datasets highlight the effectiveness of our DML method against existing state-of-the-art methods for semi-supervised text classification. We release our code and data of this paper publicly at https://github.com/GRIT621/DML.