Xinyuan Song
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xinyuan Song has a total of 69 co-authors. Here are the top co-authors by number of shared publications:
- Yan Zhong
School of Mathematical Sciences, State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University, Beijing, China
4 shared publications - Yangfan He
China West Normal University
3 shared publications - Ersheng Ni
The University of Queensland, Brisbane, Australia
2 shared publications - Hongyang He
University of Warwick
2 shared publications - Jianhui Wang
University of Electronic Science and Technology of China, Chengdu, China
2 shared publications - Li Zhang
University of Science and Technology of China, Hefei, China
2 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 15812-15821
Generalizable Blind Image Quality Assessment with Retrieval Augmentation and Contrastive Vision-Text Local Meta-Training
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 15812-15821
Open AccessGeneralizable Blind Image Quality Assessment with Retrieval Augmentation and Contrastive Vision-Text Local Meta-Training
- Open Access
- 12 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
- Vol. 58 (16) , 1-45
Beyond Efficiency: A Systematic Survey of Resource-Efficient Large Language Models
The burgeoning field of Large Language Models (LLMs), exemplified by sophisticated models like OpenAI’s ChatGPT, represents a significant advancement in artificial intelligence. These models, however, bring forth substantial challenges in high consumption of computational, memory, energy, and financial resources, especially in environments with limited resource capabilities. This survey aims to systematically address these challenges by reviewing a broad spectrum of techniques designed to enhance the resource efficiency of LLMs. We categorize methods based on their optimization focus—covering computational, memory, energy, financial, and network resources—and their applicability across various stages of an LLM’s lifecycle, including architecture design, pre-training, fine-tuning, and system design. Additionally, the survey introduces a nuanced categorization of resource efficiency techniques by their specific resource types, which uncovers the intricate relationships and mappings between various resources and corresponding optimization techniques. A standardized set of evaluation metrics and datasets is also presented to facilitate consistent and fair comparisons across different models and techniques. By offering a comprehensive overview of the current state-of-the-art and identifying open research avenues, this survey serves as a foundational reference for researchers and practitioners, aiding them in developing more sustainable and efficient LLMs in a rapidly evolving landscape. To make the field more accessible to interested beginners, we not only make a systematic review of existing works and a highly structured taxonomy of resource-efficient LLMs but also release a website including a constantly-updated article list https://github.com/tiingweii-shii/Awesome-Resource-Efficient-LLM-Papers.
- 12 October 2026
- Published by Association for Computing Machinery (ACM) in ACM Computing Surveys
- Vol. 58 (16) , 1-45
Open AccessBeyond Efficiency: A Systematic Survey of Resource-Efficient Large Language Models
The burgeoning field of Large Language Models (LLMs), exemplified by sophisticated models like OpenAI’s ChatGPT, represents a significant advancement in artificial intelligence. These models, however, bring forth substantial challenges in high consumption of computational, memory, energy, and financial resources, especially in environments with limited resource capabilities. This survey aims to systematically address these challenges by reviewing a broad spectrum of techniques designed to enhance the resource efficiency of LLMs. We categorize methods based on their optimization focus—covering computational, memory, energy, financial, and network resources—and their applicability across various stages of an LLM’s lifecycle, including architecture design, pre-training, fine-tuning, and system design. Additionally, the survey introduces a nuanced categorization of resource efficiency techniques by their specific resource types, which uncovers the intricate relationships and mappings between various resources and corresponding optimization techniques. A standardized set of evaluation metrics and datasets is also presented to facilitate consistent and fair comparisons across different models and techniques. By offering a comprehensive overview of the current state-of-the-art and identifying open research avenues, this survey serves as a foundational reference for researchers and practitioners, aiding them in developing more sustainable and efficient LLMs in a rapidly evolving landscape. To make the field more accessible to interested beginners, we not only make a systematic review of existing works and a highly structured taxonomy of resource-efficient LLMs but also release a website including a constantly-updated article list https://github.com/tiingweii-shii/Awesome-Resource-Efficient-LLM-Papers.
- 12 September 2026
- Published by Springer Nature
- p. 575-591
Geometric Regularization for Long-Tailed Semi-Supervised Learning via Gaussian Feature Bridges
- 12 September 2026
- Published by Springer Nature
- p. 575-591
Geometric Regularization for Long-Tailed Semi-Supervised Learning via Gaussian Feature Bridges
- Top Cited PapersOpen Access
- 29 May 2026
- Published by MDPI AG in Computation
- Vol. 14 (6) , 125
A Comprehensive Survey and Guide to Multimodal Large Language Models in Vision–Language Tasks
This survey provides a comprehensive guide to Multimodal Large Language Models (MLLMs) with a focus on vision–language tasks, including image captioning, visual question answering, cross-modal retrieval, visual grounding, multi-image reasoning, long-video understanding, and embodied AI. We examine architectures, training pipelines, and practical applications, covering visual encoders, language model backbones, connector modules, contrastive pre-training, instruction tuning, and preference alignment. We also foreground first-principles constraints—information bottlenecks, data-processing limits, and statistical co-occurrence bias—that shape architecture, robustness, and evaluation. This survey centers on vision–language systems and does not cover audio-only models or code-generation tools without visual inputs. Through task-level analysis and system-level case studies, we examine prominent MLLM implementations while addressing key challenges in scalability, memory, energy use, inference cost, robustness, and cross-modal learning. We present a unified taxonomy of the MLLM design space, a comparative overview of representative models and evaluation benchmarks, and a discussion of open problems. Concluding with ethical considerations and responsible AI development, this survey offers theoretical frameworks and practical insights for researchers, practitioners, and students working at the intersection of natural language processing and computer vision.
- 29 May 2026
- Published by MDPI AG in Computation
- Vol. 14 (6) , 125
Top Cited PapersOpen AccessA Comprehensive Survey and Guide to Multimodal Large Language Models in Vision–Language Tasks
This survey provides a comprehensive guide to Multimodal Large Language Models (MLLMs) with a focus on vision–language tasks, including image captioning, visual question answering, cross-modal retrieval, visual grounding, multi-image reasoning, long-video understanding, and embodied AI. We examine architectures, training pipelines, and practical applications, covering visual encoders, language model backbones, connector modules, contrastive pre-training, instruction tuning, and preference alignment. We also foreground first-principles constraints—information bottlenecks, data-processing limits, and statistical co-occurrence bias—that shape architecture, robustness, and evaluation. This survey centers on vision–language systems and does not cover audio-only models or code-generation tools without visual inputs. Through task-level analysis and system-level case studies, we examine prominent MLLM implementations while addressing key challenges in scalability, memory, energy use, inference cost, robustness, and cross-modal learning. We present a unified taxonomy of the MLLM design space, a comparative overview of representative models and evaluation benchmarks, and a discussion of open problems. Concluding with ethical considerations and responsible AI development, this survey offers theoretical frameworks and practical insights for researchers, practitioners, and students working at the intersection of natural language processing and computer vision.
- Open Access
- 16 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Complex networks of AI agentic systems: topology, memory, and update dynamics
Large-scale networks of agents are increasingly applied to software engineering, scientific analysis, web automation, organizational workflows, and social simulation, yet existing multi-agent architectures lack a unified framework to explain why some designs scale to long-horizon, multi-step tasks while others fail. As these systems grow, their behavior is fundamentally shaped by how agents are connected, how information is stored, and how states are updated over time. In this survey, we introduce a hierarchical taxonomy of agent systems along three core dimensions-architecture topology (centralized vs. decentralized), memory scope (global vs. local), and update behavior (static vs. dynamic)-which together induce eight system categories that organize prior work and make architectural trade-offs explicit. Using this taxonomy, we analyze how design choices influence scalability, coordination efficiency, communication overhead, planning depth, and robustness under partial failure, and we identify common failure modes and open challenges, including consistency management, agent routing, federation boundaries, and stability under noise or disruption.
- 16 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessComplex networks of AI agentic systems: topology, memory, and update dynamics
Large-scale networks of agents are increasingly applied to software engineering, scientific analysis, web automation, organizational workflows, and social simulation, yet existing multi-agent architectures lack a unified framework to explain why some designs scale to long-horizon, multi-step tasks while others fail. As these systems grow, their behavior is fundamentally shaped by how agents are connected, how information is stored, and how states are updated over time. In this survey, we introduce a hierarchical taxonomy of agent systems along three core dimensions-architecture topology (centralized vs. decentralized), memory scope (global vs. local), and update behavior (static vs. dynamic)-which together induce eight system categories that organize prior work and make architectural trade-offs explicit. Using this taxonomy, we analyze how design choices influence scalability, coordination efficiency, communication overhead, planning depth, and robustness under partial failure, and we identify common failure modes and open challenges, including consistency management, agent routing, federation boundaries, and stability under noise or disruption.
- Open Access
- 2 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Knowledge Discovery From Data
- Vol. 20 (2) , 1-19
Bridging NIP and MLM: A Unified Meta-Learning Framework for Sequential Recommendation
Sequential Recommender Systems (SRSs) predict items of interest for users based on their historical interactions. Two key popular paradigms for SRs are unidirectional Next Item Prediction (NIP) and bidirectional Masked Language Modeling (MLM). NIP performs well in recommendation tasks but is constrained by its reliance on prior information and a rigid temporal order assumption, limiting its ability to capture dynamic personalized preferences. MLM, on the other hand, enhances sequence representation and personalization by leveraging bidirectional context, but its objective is misaligned with recommendation tasks. To this end, we develop MetaSR, a meta-learning-based SR approach that ingeniously combines both paradigms into a unified framework to enhance personalized sequential recommendation. Specifically, we treat the mask prediction task in MLM as the inner-loop task (i.e., the support set), adjusting mask items dynamically to generate sequences that match individual behavioral patterns to capture personalized user preferences. In the outer loop (i.e., the query set), we adopt the NIP paradigm to directly train MetaSR toward the SR target. The joint training bridges the gap between MLM and SRs tasks, also overcoming NIP’s limitations in capturing dynamic personalized preferences. Furthermore, to improve the overall efficiency, we develop a reinforcement learning-based Adaptive Masked Sequence Selection (AMSS) mechanism, automatically selecting the optimal masking prediction tasks within the meta-learning support set to accelerate the meta-learning and achieve better personalization. Experiments on multiple public benchmark datasets show that MetaSR significantly outperforms existing SRs models. The dataset and codes are available at MetaSR .
- 2 February 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Knowledge Discovery From Data
- Vol. 20 (2) , 1-19
Open AccessBridging NIP and MLM: A Unified Meta-Learning Framework for Sequential Recommendation
Sequential Recommender Systems (SRSs) predict items of interest for users based on their historical interactions. Two key popular paradigms for SRs are unidirectional Next Item Prediction (NIP) and bidirectional Masked Language Modeling (MLM). NIP performs well in recommendation tasks but is constrained by its reliance on prior information and a rigid temporal order assumption, limiting its ability to capture dynamic personalized preferences. MLM, on the other hand, enhances sequence representation and personalization by leveraging bidirectional context, but its objective is misaligned with recommendation tasks. To this end, we develop MetaSR, a meta-learning-based SR approach that ingeniously combines both paradigms into a unified framework to enhance personalized sequential recommendation. Specifically, we treat the mask prediction task in MLM as the inner-loop task (i.e., the support set), adjusting mask items dynamically to generate sequences that match individual behavioral patterns to capture personalized user preferences. In the outer loop (i.e., the query set), we adopt the NIP paradigm to directly train MetaSR toward the SR target. The joint training bridges the gap between MLM and SRs tasks, also overcoming NIP’s limitations in capturing dynamic personalized preferences. Furthermore, to improve the overall efficiency, we develop a reinforcement learning-based Adaptive Masked Sequence Selection (AMSS) mechanism, automatically selecting the optimal masking prediction tasks within the meta-learning support set to accelerate the meta-learning and achieve better personalization. Experiments on multiple public benchmark datasets show that MetaSR significantly outperforms existing SRs models. The dataset and codes are available at MetaSR .
- Open Access
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 2801-2810
AMBER: Adaptive Meta Balanced Paradigm for Heterogeneous Graph-Based Knowledge Tracing
- 10 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 2801-2810
Open AccessAMBER: Adaptive Meta Balanced Paradigm for Heterogeneous Graph-Based Knowledge Tracing
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 7453-7462
Adaptive Prompt Learning for Blind Image Quality Assessment with Multi-modal Mixed-datasets Training
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 7453-7462
Open AccessAdaptive Prompt Learning for Blind Image Quality Assessment with Multi-modal Mixed-datasets Training
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9881-9890
Free-Mask: A Novel Paradigm of Integration Between the Segmentation Diffusion Model and Image Editing
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9881-9890
Open AccessFree-Mask: A Novel Paradigm of Integration Between the Segmentation Diffusion Model and Image Editing
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 3645-3653
Twin Co-Adaptive Dialogue for Progressive Image Generation
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 3645-3653
Open AccessTwin Co-Adaptive Dialogue for Progressive Image Generation