Sichun Luo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Sichun Luo has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Linqi Song
City University of Hong Kong
9 shared publications - Jian Xu
Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China
4 shared publications - Yuanzhang Xiao
University of Hawaii at Manoa, Honolulu, HI, USA
4 shared publications - Yuxuan Yao
City University of Hong Kong, Hong Kong
4 shared publications - Hanxu Hou
Dongguan University of Technology, Dongguan, China
3 shared publications - Haohan Zhao
Chinese Academy of Sciences, Hong Kong Institute of Science & Innovation, Hong Kong, Hong Kong
3 shared publications
Latest Publications
- Top Cited Papers
- 16 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 20 (1) , 77-89
Integrating Large Language Models Into Recommendation via Mutual Augmentation and Adaptive Aggregation
Conventional recommender systems and Large Language Model (LLM)-based recommender systems each have their strengths and weaknesses. While conventional recommendation methods excel at mining collaborative information and modeling sequential behavior, they struggle with data sparsity and the long-tail problem. LLM, on the other hand, is proficient at utilizing rich textual contexts but faces challenges in mining collaborative or sequential information. Despite their individual successes, there is a significant gap in leveraging their ensemble potential to enhance recommendation performance. In this paper, we introduce a general and model-agnostic framework known as Large language models with mutual augmentation and adaptive aggregation for Recommendation (Llama4Rec), aiming to bridge this gap via explicitly ensemble LLM and conventional recommendation model for more effective recommendation. We propose data augmentation and prompt augmentation strategies tailored to enhance the conventional recommendation model and LLM respectively. An adaptive aggregation module is adopted to combine the predictions of both kinds of models to refine the final recommendation results. Empirical studies on three datasets validate the superiority of Llama4Rec, demonstrating significant improvements in recommendation performance.
- 16 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 20 (1) , 77-89
Top Cited PapersIntegrating Large Language Models Into Recommendation via Mutual Augmentation and Adaptive Aggregation
Conventional recommender systems and Large Language Model (LLM)-based recommender systems each have their strengths and weaknesses. While conventional recommendation methods excel at mining collaborative information and modeling sequential behavior, they struggle with data sparsity and the long-tail problem. LLM, on the other hand, is proficient at utilizing rich textual contexts but faces challenges in mining collaborative or sequential information. Despite their individual successes, there is a significant gap in leveraging their ensemble potential to enhance recommendation performance. In this paper, we introduce a general and model-agnostic framework known as Large language models with mutual augmentation and adaptive aggregation for Recommendation (Llama4Rec), aiming to bridge this gap via explicitly ensemble LLM and conventional recommendation model for more effective recommendation. We propose data augmentation and prompt augmentation strategies tailored to enhance the conventional recommendation model and LLM respectively. An adaptive aggregation module is adopted to combine the predictions of both kinds of models to refine the final recommendation results. Empirical studies on three datasets validate the superiority of Llama4Rec, demonstrating significant improvements in recommendation performance.
- 29 August 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 297
RALLRec+: Retrieval augmented large language model recommendation with reasoning
- 29 August 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 297
RALLRec+: Retrieval augmented large language model recommendation with reasoning
- Top Cited PapersOpen Access
- 10 July 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 43 (5) , 1-31
RecRanker: Instruction Tuning Large Language Model as Ranker for Top-k Recommendation
Large language models (LLMs) have demonstrated remarkable capabilities and have been extensively deployed across various domains, including recommender systems. Prior research has employed specialized prompts to leverage the in-context learning capabilities of LLMs for recommendation purposes. More recent studies have utilized instruction tuning techniques to align LLMs with human preferences, promising more effective recommendations. However, existing methods suffer from several limitations. The full potential of LLMs is not fully elicited due to low-quality tuning data and the overlooked integration of conventional recommender signals. Furthermore, LLMs may generate inconsistent responses for different ranking tasks in the recommendation, potentially leading to unreliable results. In this article, we introduce Ranker for top- k Recommendations (RecRanker), tailored for instruction tuning LLMs to serve as the Ranker for top- k Recommendations. Specifically, we introduce importance-aware sampling, clustering-based sampling, and penalty for repetitive sampling for sampling high-quality, representative, and diverse training data. To enhance the prompt, we introduce a position shifting strategy to mitigate position bias and augment the prompt with auxiliary information from conventional recommendation models, thereby enriching the contextual understanding of the LLM. Subsequently, we utilize the sampled data to assemble an instruction-tuning dataset with the augmented prompts comprising three distinct ranking tasks: pointwise, pairwise, and listwise rankings. We further propose a hybrid ranking method to enhance the model performance by ensembling these ranking tasks. Our empirical evaluations demonstrate the effectiveness of our proposed RecRanker in both direct and sequential recommendation scenarios. 1
- 10 July 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 43 (5) , 1-31
Top Cited PapersOpen AccessRecRanker: Instruction Tuning Large Language Model as Ranker for Top-k Recommendation
Large language models (LLMs) have demonstrated remarkable capabilities and have been extensively deployed across various domains, including recommender systems. Prior research has employed specialized prompts to leverage the in-context learning capabilities of LLMs for recommendation purposes. More recent studies have utilized instruction tuning techniques to align LLMs with human preferences, promising more effective recommendations. However, existing methods suffer from several limitations. The full potential of LLMs is not fully elicited due to low-quality tuning data and the overlooked integration of conventional recommender signals. Furthermore, LLMs may generate inconsistent responses for different ranking tasks in the recommendation, potentially leading to unreliable results. In this article, we introduce Ranker for top- k Recommendations (RecRanker), tailored for instruction tuning LLMs to serve as the Ranker for top- k Recommendations. Specifically, we introduce importance-aware sampling, clustering-based sampling, and penalty for repetitive sampling for sampling high-quality, representative, and diverse training data. To enhance the prompt, we introduce a position shifting strategy to mitigate position bias and augment the prompt with auxiliary information from conventional recommendation models, thereby enriching the contextual understanding of the LLM. Subsequently, we utilize the sampled data to assemble an instruction-tuning dataset with the augmented prompts comprising three distinct ranking tasks: pointwise, pairwise, and listwise rankings. We further propose a hybrid ranking method to enhance the model performance by ensembling these ranking tasks. Our empirical evaluations demonstrate the effectiveness of our proposed RecRanker in both direct and sequential recommendation scenarios. 1
- Open Access
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 1436-1440
RALLRec: Improving Retrieval Augmented Large Language Model Recommendation with Representation Learning
- 8 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 1436-1440
Open AccessRALLRec: Improving Retrieval Augmented Large Language Model Recommendation with Representation Learning
- 13 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 19 (1) , 195-207
Cooperative Multi-Model Training for Personalized Federated Learning Over Heterogeneous Devices
Federated learning (FL) is an increasingly popular paradigm for protecting data privacy in machine learning systems. However, the data heterogeneity and high computation cost/latency are challenging barriers for employing FL in real-world applications with heterogeneous devices. In this paper, we propose a novel personalized FL framework named $\mathtt {CompFL}$ allowing cooperative training of models with varied structures to mitigate those issues. First, $\mathtt {CompFL}$ initializes a set of expert models in varied sizes and allows each client to choose one or multiple expert models for training according to their capacities. Second, $\mathtt {CompFL}$ combines the model decoupling strategy and local-global feature alignment to mitigate the adverse impact of label heterogeneity, where clients only share the feature extractor part for each model architecture. Third, to encourage mutual enhancement of various models, knowledge distillation in local training is further applied to improve the overall performance. To make our framework workable in real systems, we implement it in both centralized settings with server-coordinated parallel training, and decentralized settings with newly developed device-to-device training-forwarding schemes. Extensive experiments on benchmark datasets are conducted to verify the potential of our framework for personalized FL over heterogeneous devices.
- 13 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 19 (1) , 195-207
Cooperative Multi-Model Training for Personalized Federated Learning Over Heterogeneous Devices
Federated learning (FL) is an increasingly popular paradigm for protecting data privacy in machine learning systems. However, the data heterogeneity and high computation cost/latency are challenging barriers for employing FL in real-world applications with heterogeneous devices. In this paper, we propose a novel personalized FL framework named $\mathtt {CompFL}$ allowing cooperative training of models with varied structures to mitigate those issues. First, $\mathtt {CompFL}$ initializes a set of expert models in varied sizes and allows each client to choose one or multiple expert models for training according to their capacities. Second, $\mathtt {CompFL}$ combines the model decoupling strategy and local-global feature alignment to mitigate the adverse impact of label heterogeneity, where clients only share the feature extractor part for each model architecture. Third, to encourage mutual enhancement of various models, knowledge distillation in local training is further applied to improve the overall performance. To make our framework workable in real systems, we implement it in both centralized settings with server-coordinated parallel training, and decentralized settings with newly developed device-to-device training-forwarding schemes. Extensive experiments on benchmark datasets are conducted to verify the potential of our framework for personalized FL over heterogeneous devices.
- Open Access
- 31 October 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
- Vol. 15 (5) , 1-24
PerFedRec++: Enhancing Personalized Federated Recommendation with Self-Supervised Pre-Training
Federated recommendation systems employ federated learning techniques to safeguard user privacy by transmitting model parameters instead of raw user data between user devices and the central server. Nevertheless, the current federated recommender system faces three significant challenges: (1) data heterogeneity: the heterogeneity of users’ attributes and local data necessitates the acquisition of personalized models to improve the performance of federated recommendation; (2) model performance degradation: the privacy-preserving protocol design in the federated recommendation, such as pseudo item labeling and differential privacy, would deteriorate the model performance; (3) communication bottleneck: the standard federated recommendation algorithm can have a high communication overhead. Previous studies have attempted to address these issues, but none have been able to solve them simultaneously. In this article, we propose a novel framework, named PerFedRec++ , to enhance the personalized federated recommendation with self-supervised pre-training. Specifically, we utilize the privacy-preserving mechanism of federated recommender systems to generate two augmented graph views, which are used as contrastive tasks in self-supervised graph learning to pre-train the model. Pre-training enhances the performance of federated models by improving the uniformity of representation learning. Also, by providing a better initial state for federated training, pre-training makes the overall training converge faster, thus alleviating the heavy communication burden. We then construct a collaborative graph to learn the client representation through a federated graph neural network. Based on these learned representations, we cluster users into different user groups and learn personalized models for each cluster. Each user learns a personalized model by combining the global federated model, the cluster-level federated model, and its own fine-tuned local model. Experiments on three real-world datasets show that our proposed method achieves superior performance over existing methods.
- 31 October 2024
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
- Vol. 15 (5) , 1-24
Open AccessPerFedRec++: Enhancing Personalized Federated Recommendation with Self-Supervised Pre-Training
Federated recommendation systems employ federated learning techniques to safeguard user privacy by transmitting model parameters instead of raw user data between user devices and the central server. Nevertheless, the current federated recommender system faces three significant challenges: (1) data heterogeneity: the heterogeneity of users’ attributes and local data necessitates the acquisition of personalized models to improve the performance of federated recommendation; (2) model performance degradation: the privacy-preserving protocol design in the federated recommendation, such as pseudo item labeling and differential privacy, would deteriorate the model performance; (3) communication bottleneck: the standard federated recommendation algorithm can have a high communication overhead. Previous studies have attempted to address these issues, but none have been able to solve them simultaneously. In this article, we propose a novel framework, named PerFedRec++ , to enhance the personalized federated recommendation with self-supervised pre-training. Specifically, we utilize the privacy-preserving mechanism of federated recommender systems to generate two augmented graph views, which are used as contrastive tasks in self-supervised graph learning to pre-train the model. Pre-training enhances the performance of federated models by improving the uniformity of representation learning. Also, by providing a better initial state for federated training, pre-training makes the overall training converge faster, thus alleviating the heavy communication burden. We then construct a collaborative graph to learn the client representation through a federated graph neural network. Based on these learned representations, we cluster users into different user groups and learn personalized models for each cluster. Each user learns a personalized model by combining the global federated model, the cluster-level federated model, and its own fine-tuned local model. Experiments on three real-world datasets show that our proposed method achieves superior performance over existing methods.
- Open Access
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1099-1102
Can LLM Substitute Human Labeling? A Case Study of Fine-grained Chinese Address Entity Recognition Dataset for UAV Delivery
- 13 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1099-1102
Open AccessCan LLM Substitute Human Labeling? A Case Study of Fine-grained Chinese Address Entity Recognition Dataset for UAV Delivery
- 3 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 18 (3) , 529-539
A Language Model-Based Fine-Grained Address Resolution Framework in UAV Delivery System
Accurate address resolution plays a vital role in UAV delivery systems. Existing address resolution systems heavily rely on user-provided Point of Interest (POI) information. However, such information often lacks precision, making it challenging to obtain fine-grained details for further processing. In this paper, we present an end-to-end L anguage M odel-based fine-grained A ddress R esolution framework ( LMAR ). Instead of solely relying on POI information, we introduce a language model to process the user input text information. Specifically, we start by collecting data and constructing two datasets, which are then used to fine-tune a pre-trained language model. Additionally, our pipeline incorporates pre-processing and post-processing modules to handle data processing and regularization. We combine the output of the language model with the POI information to perform a database match and derive the final outcome. To evaluate our proposed LMAR, we conduct offline and online experiments. In both offline and online testing, our proposed model achieves an overall performance of over 90% accuracy, while in the online pressure test, it achieves satisfactory performance, demonstrating its effectiveness and practicality. The proposed LMAR has passed the internal test and will be deployed into the Meituan UAV delivery system in the near future.
- 3 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Signal Processing
- Vol. 18 (3) , 529-539
A Language Model-Based Fine-Grained Address Resolution Framework in UAV Delivery System
Accurate address resolution plays a vital role in UAV delivery systems. Existing address resolution systems heavily rely on user-provided Point of Interest (POI) information. However, such information often lacks precision, making it challenging to obtain fine-grained details for further processing. In this paper, we present an end-to-end L anguage M odel-based fine-grained A ddress R esolution framework ( LMAR ). Instead of solely relying on POI information, we introduce a language model to process the user input text information. Specifically, we start by collecting data and constructing two datasets, which are then used to fine-tune a pre-trained language model. Additionally, our pipeline incorporates pre-processing and post-processing modules to handle data processing and regularization. We combine the output of the language model with the POI information to perform a database match and derive the final outcome. To evaluate our proposed LMAR, we conduct offline and online experiments. In both offline and online testing, our proposed model achieves an overall performance of over 90% accuracy, while in the online pressure test, it achieves satisfactory performance, demonstrating its effectiveness and practicality. The proposed LMAR has passed the internal test and will be deployed into the Meituan UAV delivery system in the near future.
- Open Access
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1707-1716
Improving Long-Tail Item Recommendation with Graph Augmentation
The ubiquitous long-tail distribution of inherent user behaviors results in worse recommendation performance for the items with fewer user records (i.e., tail items) than those with richer ones (i.e., head items). Graph-based recommendation methods (e.g., using graph neural networks) have recently emerged as a powerful tool for recommender systems, often outperforming traditional methods. However, existing techniques for alleviating the long-tail problem mainly focus on traditional methods. There is a lack of graph-based methods that can efficiently deal with the long-tail problem. In this paper, we propose a novel approach, Graph Augmentation for Long-tail Recommendation (GALORE), which can be plugged into any graph-based recommendation models to improve the performance for tail items. GALORE incorporates an edge addition module that enriches the graph's connectivity for tail items by injecting additional item-to-item edges. To further balance the graph structure, GALORE utilizes a degree-aware edge dropping strategy, preserving the more valuable edges from the tail items while selectively discarding less informative edges from the head items. Beyond structural augmentation, we synthesize new data samples, thereby addressing the data scarcity issue for tail items. We further introduce a two-stage training strategy to facilitate the learning for both head and tail items. Comprehensive empirical studies conducted on four datasets show that GALORE outperforms existing methods in terms of the performance for tail items as well as the overall performance.
- 21 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1707-1716
Open AccessImproving Long-Tail Item Recommendation with Graph Augmentation
The ubiquitous long-tail distribution of inherent user behaviors results in worse recommendation performance for the items with fewer user records (i.e., tail items) than those with richer ones (i.e., head items). Graph-based recommendation methods (e.g., using graph neural networks) have recently emerged as a powerful tool for recommender systems, often outperforming traditional methods. However, existing techniques for alleviating the long-tail problem mainly focus on traditional methods. There is a lack of graph-based methods that can efficiently deal with the long-tail problem. In this paper, we propose a novel approach, Graph Augmentation for Long-tail Recommendation (GALORE), which can be plugged into any graph-based recommendation models to improve the performance for tail items. GALORE incorporates an edge addition module that enriches the graph's connectivity for tail items by injecting additional item-to-item edges. To further balance the graph structure, GALORE utilizes a degree-aware edge dropping strategy, preserving the more valuable edges from the tail items while selectively discarding less informative edges from the head items. Beyond structural augmentation, we synthesize new data samples, thereby addressing the data scarcity issue for tail items. We further introduce a two-stage training strategy to facilitate the learning for both head and tail items. Comprehensive empirical studies conducted on four datasets show that GALORE outperforms existing methods in terms of the performance for tail items as well as the overall performance.