Tong Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Tong Zhang has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Guandong Xu
The Education University of Hong Kong
4 shared publications - Xianzhi Wang
University of Technology Sydney, Sydney, Australia
4 shared publications - Zihao Li
University of Technology Sydney, Sydney, Australia
4 shared publications - Yakun Chen
Center for Learning, Teaching and Technology, The Education University of Hong Kong, Hong Kong, 999077, Hong Kong, China
2 shared publications - Chao Yang
Faculty of Information Science and Engineering, Ocean University of China, Qingdao, China
1 shared publication - Chongjun Xia
Chinese Academy of Sciences Chongqing Institute of Green and Intelligent Technology
1 shared publication
Latest Publications
- Open Access
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 6300-6310
SCAU-REC: Subspace-Constrained Adapter Updates for LLM-Based Recommendations
- 6 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 6300-6310
Open AccessSCAU-REC: Subspace-Constrained Adapter Updates for LLM-Based Recommendations
- Open Access
- 5 March 2026
- Published by Elsevier in SSRN Electronic Journal
NeighCF: Neighborhood-aware Collaborative Filtering for LLM-based Sequential Recommendation
Recently, Large Language Models (LLMs) have shown strong promise in recommendation tasks by leveraging semantic reasoning and external knowledge. As inputs to LLMs, prompts play a crucial role in guiding LLMs to generate outputs that align with task-specific requirements. However, existing prompting strategies primarily rely on a user’s individual history, overlooking collaborative signals that capture broader behavioral patterns. While recent works attempt to incorporate neighbors’ histories, they face two key challenges: (i) fixed-size neighbor selection, which either underutilizes sparse contexts or overloads dense ones with redundancy; and (ii) computationally expensive global neighbor retrieval, which is impractical for large-scale systems. Specifically, NeighCF (i) employs a cluster-aware retrieval strategy that restricts neighbor search to candidate-relevant subspaces, reducing the cost of global retrieval while maintaining high recall; and (ii) introduces an adaptive neighbor selection mechanism based on a learnable threshold that dynamically adjusts the number of neighbors according to interaction density and candidate relevance. The resulting neighbor representations are integrated into prompts for LLM-based reasoning. Extensive experiments on various datasets demonstrate that NeighCF consistently outperforms both single-user prompting baselines and state-of-the-art collaborative prompting approaches. These results validate the effectiveness and efficiency of our framework, highlighting collaborative reasoning in LLM-based sequential recommendation.
- 5 March 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessNeighCF: Neighborhood-aware Collaborative Filtering for LLM-based Sequential Recommendation
Recently, Large Language Models (LLMs) have shown strong promise in recommendation tasks by leveraging semantic reasoning and external knowledge. As inputs to LLMs, prompts play a crucial role in guiding LLMs to generate outputs that align with task-specific requirements. However, existing prompting strategies primarily rely on a user’s individual history, overlooking collaborative signals that capture broader behavioral patterns. While recent works attempt to incorporate neighbors’ histories, they face two key challenges: (i) fixed-size neighbor selection, which either underutilizes sparse contexts or overloads dense ones with redundancy; and (ii) computationally expensive global neighbor retrieval, which is impractical for large-scale systems. Specifically, NeighCF (i) employs a cluster-aware retrieval strategy that restricts neighbor search to candidate-relevant subspaces, reducing the cost of global retrieval while maintaining high recall; and (ii) introduces an adaptive neighbor selection mechanism based on a learnable threshold that dynamically adjusts the number of neighbors according to interaction density and candidate relevance. The resulting neighbor representations are integrated into prompts for LLM-based reasoning. Extensive experiments on various datasets demonstrate that NeighCF consistently outperforms both single-user prompting baselines and state-of-the-art collaborative prompting approaches. These results validate the effectiveness and efficiency of our framework, highlighting collaborative reasoning in LLM-based sequential recommendation.
- 26 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. PP, 1-14
Listwise Preference Alignment Optimization for Tail Item Recommendation
Preference alignment has achieved greater success on large language models (LLMs) and drawn broad interest in recommendation research. Existing preference alignment methods for recommendation either require explicit reward modeling or only support pairwise preference comparison. The former directly increases substantial computational costs, while the latter hinders training efficiency on negative samples. Moreover, no existing effort has explored preference alignment solutions for tail-item recommendation. To bridge the above gaps, we propose LPO4Rec, which extends the Bradley-Terry model from pairwise comparison to listwise comparison, to improve the efficiency of model training. Specifically, we derive a closed-form optimal policy to enable more efficient and effective training without explicit reward modeling. We also present an adaptive negative sampling and reweighting strategy to prioritize tail items during optimization and enhance performance in tail-item recommendations. Besides, we theoretically prove that optimizing the listwise preference optimization (LPO) loss is equivalent to maximizing the upper bound of the optimal reward. Our experiments on three public datasets show that our method outperforms 10 baselines by a large margin, achieving up to 50% performance improvement while reducing 17.9% GPU memory usage when compared with direct preference optimization (DPO) in tail-item recommendation.
- 26 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. PP, 1-14
Listwise Preference Alignment Optimization for Tail Item Recommendation
Preference alignment has achieved greater success on large language models (LLMs) and drawn broad interest in recommendation research. Existing preference alignment methods for recommendation either require explicit reward modeling or only support pairwise preference comparison. The former directly increases substantial computational costs, while the latter hinders training efficiency on negative samples. Moreover, no existing effort has explored preference alignment solutions for tail-item recommendation. To bridge the above gaps, we propose LPO4Rec, which extends the Bradley-Terry model from pairwise comparison to listwise comparison, to improve the efficiency of model training. Specifically, we derive a closed-form optimal policy to enable more efficient and effective training without explicit reward modeling. We also present an adaptive negative sampling and reweighting strategy to prioritize tail items during optimization and enhance performance in tail-item recommendations. Besides, we theoretically prove that optimizing the listwise preference optimization (LPO) loss is equivalent to maximizing the upper bound of the optimal reward. Our experiments on three public datasets show that our method outperforms 10 baselines by a large margin, achieving up to 50% performance improvement while reducing 17.9% GPU memory usage when compared with direct preference optimization (DPO) in tail-item recommendation.
- Open Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 4181-4190
SarRec: Statistically-guaranteed Augmented Retrieval for Recommendation
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 4181-4190
Open AccessSarRec: Statistically-guaranteed Augmented Retrieval for Recommendation
- Open Access
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 4925-4936
Reembedding and Reweighting are Needed for Tail Item Sequential Recommendation
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 4925-4936
Open AccessReembedding and Reweighting are Needed for Tail Item Sequential Recommendation
- RetractedOpen Access
- 30 January 2022
- Published by Hindawi Limited in Computational Intelligence and Neuroscience
- Vol. 2022, 1-10
Design of a Regional Economic Forecasting Model Using Optimal Nonlinear Support Vector Machines
Forecasting regional economic activity is a progressively significant element of regional economic research. Regional economic prediction can directly assist local, national, and subnational policymakers. Regional economic activity forecast can be employed for defining macroeconomic forces, such as prediction of stock market and cyclicality of national labor market movement. The recent advances of machine learning (ML) models can be employed to solve the time series prediction problem. Since the parameters involved in the ML model considerably influence the performance, the parameter tuning process also becomes essential. With this motivation, this study develops a quasioppositional cuckoo search algorithm (QOCSA) with a nonlinear support vector machine (SVM)-based prediction model, called QOCSO-NLSVM for regional economic prediction. The goal of the QOCSO-NLSVM technique is to identify the present regional economic status. The QOCSO-NLSVM technique has different stages such as clustering, preprocessing, prediction, and optimization. Besides, the QOCSO-NLSVM technique employs the density-based clustering algorithm (DBSCAN) to determine identical states depending upon the per capita NSDP growth trends and socio-economic-demographic features in a state. Moreover, the NLSVM model is employed for the time series prediction process and the parameters involved in it are optimally tuned by the use of the QOCSO algorithm. To showcase the effective performance of the QOCSO-NLSVM technique, a wide range of simulations take place using regional economic data. To determine the current economic situation in a region, the QOCSO-NLSVM technique is used. The simulation results reported the better performance of the QOCSO-NLSVM technique over recent approaches. The QOCSO-NLSVM technique generated effective results with a minimal mean square error of 70.548 or greater. Astonishingly good results were obtained using the QOCSO-NLSVM approach, which had the lowest root mean square error (RMSE) of 8.399.
- 30 January 2022
- Published by Hindawi Limited in Computational Intelligence and Neuroscience
- Vol. 2022, 1-10
RetractedOpen AccessDesign of a Regional Economic Forecasting Model Using Optimal Nonlinear Support Vector Machines
Forecasting regional economic activity is a progressively significant element of regional economic research. Regional economic prediction can directly assist local, national, and subnational policymakers. Regional economic activity forecast can be employed for defining macroeconomic forces, such as prediction of stock market and cyclicality of national labor market movement. The recent advances of machine learning (ML) models can be employed to solve the time series prediction problem. Since the parameters involved in the ML model considerably influence the performance, the parameter tuning process also becomes essential. With this motivation, this study develops a quasioppositional cuckoo search algorithm (QOCSA) with a nonlinear support vector machine (SVM)-based prediction model, called QOCSO-NLSVM for regional economic prediction. The goal of the QOCSO-NLSVM technique is to identify the present regional economic status. The QOCSO-NLSVM technique has different stages such as clustering, preprocessing, prediction, and optimization. Besides, the QOCSO-NLSVM technique employs the density-based clustering algorithm (DBSCAN) to determine identical states depending upon the per capita NSDP growth trends and socio-economic-demographic features in a state. Moreover, the NLSVM model is employed for the time series prediction process and the parameters involved in it are optimally tuned by the use of the QOCSO algorithm. To showcase the effective performance of the QOCSO-NLSVM technique, a wide range of simulations take place using regional economic data. To determine the current economic situation in a region, the QOCSO-NLSVM technique is used. The simulation results reported the better performance of the QOCSO-NLSVM technique over recent approaches. The QOCSO-NLSVM technique generated effective results with a minimal mean square error of 70.548 or greater. Astonishingly good results were obtained using the QOCSO-NLSVM approach, which had the lowest root mean square error (RMSE) of 8.399.