Yaru Zhao
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
Yaru Zhao has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Yakun Huang
State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing, China
6 shared publications - Bo Cheng
School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
4 shared publications - Yi Yue
China Unicom Research Institute, Beijing, China
3 shared publications - Xiuquan Qiao
State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing, China
2 shared publications Baojun Ma
China Unicom Online Information Technology Company Ltd., Beijing, China
1 shared publicationChang Cao
China Unicom Research Institute
1 shared publication
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (11) , 19659-19674
AvailGuard: An Availability-Aware VNF Deployment Framework Based on Traffic Forecasting in SDN/NFV-Enabled Networks
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (11) , 19659-19674
AvailGuard: An Availability-Aware VNF Deployment Framework Based on Traffic Forecasting in SDN/NFV-Enabled Networks
- 3 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 23, 3850-3864
EcoPath: Energy-Efficient Multi-Path Data Aggregation for Ubiquitous Connectivity Services
Ubiquitous connectivity is a key 6G usage scenario, in which large-scale sensing systems deployed in remote and underserved regions must deliver heterogeneous sensing data under stringent energy budgets and deadline constraints. This paper presents EcoPath, a two-tier data aggregation framework for clustered large-scale sensor networks. EcoPath separates low-power intra-cluster collection from a high-rate multi-interface backhaul operated by cluster heads, where Multipath QUIC (MPQUIC) can be practically deployed to exploit path diversity. At the cluster head, EcoPath jointly integrates (i) a deadline-aware bundling controller that aggregates sensor frames into MTU-bounded bundles to amortize protocol overhead while bounding additional waiting time, and (ii) a robust multi-path scheduler that prioritizes packets using Weighted Earliest- Deadline-First (W-EDF) with fairness protection and selects backhaul paths via a stability-aware quality metric with hysteresis to avoid flapping under time-varying links. We further formulate an explicit energy–timeliness optimization and show how its outputs parameterize the online bundling and scheduling policies. Extensive simulations with realistic wireless effects, together with baselines and ablations, demonstrate that EcoPath improves energy efficiency and deadline satisfaction for large-scale aggregation.
- 3 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network and Service Management
- Vol. 23, 3850-3864
EcoPath: Energy-Efficient Multi-Path Data Aggregation for Ubiquitous Connectivity Services
Ubiquitous connectivity is a key 6G usage scenario, in which large-scale sensing systems deployed in remote and underserved regions must deliver heterogeneous sensing data under stringent energy budgets and deadline constraints. This paper presents EcoPath, a two-tier data aggregation framework for clustered large-scale sensor networks. EcoPath separates low-power intra-cluster collection from a high-rate multi-interface backhaul operated by cluster heads, where Multipath QUIC (MPQUIC) can be practically deployed to exploit path diversity. At the cluster head, EcoPath jointly integrates (i) a deadline-aware bundling controller that aggregates sensor frames into MTU-bounded bundles to amortize protocol overhead while bounding additional waiting time, and (ii) a robust multi-path scheduler that prioritizes packets using Weighted Earliest- Deadline-First (W-EDF) with fairness protection and selects backhaul paths via a stability-aware quality metric with hysteresis to avoid flapping under time-varying links. We further formulate an explicit energy–timeliness optimization and show how its outputs parameterize the online bundling and scheduling policies. Extensive simulations with realistic wireless effects, together with baselines and ablations, demonstrate that EcoPath improves energy efficiency and deadline satisfaction for large-scale aggregation.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (6) , 12557-12571
Position Encoding-Enhanced Adaptive Expansion Graph Attention Network for Anomaly Detection in Multivariate Time Series
As cyber–physical systems continue to increase in complexity, multivariate time series exhibit not only intricate temporal patterns within individual variables but also complex inter-variable dependencies, including both synchronous and asynchronous propagation. Although existing Graph Neural Network (GNN)-based methods perform well in modeling variable dependencies, they often decouple temporal pattern processing from the graph construction. This separation often results in graph structures that lack temporal semantics and struggle to capture delayed responses among variables. To address this, we propose a Position Encoding–Enhanced Adaptive Expansion Graph Attention Network (PEE-AEGAT) for multivariate time series anomaly detection and root cause analysis. First, to embed temporal semantics directly into the graph construction process, we design a Multi-Frequency Positional Encoding Projection (MF-PEP) module that explicitly injects multi-scale periodic patterns and positional information into node embeddings. Second, we introduce a Hybrid Adaptive Graph Structure Learning (HAGSL) strategy that constructs a dual-mode graph by integrating static feature similarity with dynamic delay-aware similarity, together with a density-aware adaptive expansion mechanism for flexibly capturing dependencies. Finally, we develop a delay-aware graph attention network for information aggregation and propose an influence-based scoring mechanism to accurately identify anomaly root causes. Extensive experiments on four public real-world datasets demonstrate that PEE-AEGAT significantly outperforms state-of-the-art baselines in anomaly detection accuracy and root cause localization performance.
- 12 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13 (6) , 12557-12571
Position Encoding-Enhanced Adaptive Expansion Graph Attention Network for Anomaly Detection in Multivariate Time Series
As cyber–physical systems continue to increase in complexity, multivariate time series exhibit not only intricate temporal patterns within individual variables but also complex inter-variable dependencies, including both synchronous and asynchronous propagation. Although existing Graph Neural Network (GNN)-based methods perform well in modeling variable dependencies, they often decouple temporal pattern processing from the graph construction. This separation often results in graph structures that lack temporal semantics and struggle to capture delayed responses among variables. To address this, we propose a Position Encoding–Enhanced Adaptive Expansion Graph Attention Network (PEE-AEGAT) for multivariate time series anomaly detection and root cause analysis. First, to embed temporal semantics directly into the graph construction process, we design a Multi-Frequency Positional Encoding Projection (MF-PEP) module that explicitly injects multi-scale periodic patterns and positional information into node embeddings. Second, we introduce a Hybrid Adaptive Graph Structure Learning (HAGSL) strategy that constructs a dual-mode graph by integrating static feature similarity with dynamic delay-aware similarity, together with a density-aware adaptive expansion mechanism for flexibly capturing dependencies. Finally, we develop a delay-aware graph attention network for information aggregation and propose an influence-based scoring mechanism to accurately identify anomaly root causes. Extensive experiments on four public real-world datasets demonstrate that PEE-AEGAT significantly outperforms state-of-the-art baselines in anomaly detection accuracy and root cause localization performance.
- 30 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 44 (07338716) , 3137-3151
Intent-Driven Cognitive XR Networks: Multi-Agent Orchestration for Immersive Communication
Extended Reality (XR) is emerging as a central use case in the 6G era, requiring intelligent, adaptive, and low-latency communication to support immersive user experiences. However, existing network architectures remain reactive and lack the capability to interpret or act upon users’ fine-grained multimodal intents, limiting their responsiveness and efficiency. To address this gap, this paper presents the Cognitive XR Network (CXN) framework, an intent-driven architecture that integrates perception, reasoning, and control through a hierarchy of Artificial Intelligence (AI) agents. CXN comprises three cooperating agents: the User Intent Agent, which infers and predicts user intents from multimodal sensory inputs; the Network Orchestration Agent, which performs global coordination through multi-agent reinforcement learning; and distributed Resource Management Agents which execute localized decisions under strategic guidance. Together, these agents form a closed cognitive loop that enables proactive, intent-aware orchestration across radio, compute, and caching domains. Simulations under realistic XR collaboration scenarios demonstrate that CXN sustains an intent satisfaction rate above 85% and an average interaction latency below 60 ms with 90 concurrent users, outperforming both reactive and centralized learning baselines.
- 30 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 44 (07338716) , 3137-3151
Intent-Driven Cognitive XR Networks: Multi-Agent Orchestration for Immersive Communication
Extended Reality (XR) is emerging as a central use case in the 6G era, requiring intelligent, adaptive, and low-latency communication to support immersive user experiences. However, existing network architectures remain reactive and lack the capability to interpret or act upon users’ fine-grained multimodal intents, limiting their responsiveness and efficiency. To address this gap, this paper presents the Cognitive XR Network (CXN) framework, an intent-driven architecture that integrates perception, reasoning, and control through a hierarchy of Artificial Intelligence (AI) agents. CXN comprises three cooperating agents: the User Intent Agent, which infers and predicts user intents from multimodal sensory inputs; the Network Orchestration Agent, which performs global coordination through multi-agent reinforcement learning; and distributed Resource Management Agents which execute localized decisions under strategic guidance. Together, these agents form a closed cognitive loop that enables proactive, intent-aware orchestration across radio, compute, and caching domains. Simulations under realistic XR collaboration scenarios demonstrate that CXN sustains an intent satisfaction rate above 85% and an average interaction latency below 60 ms with 90 concurrent users, outperforming both reactive and centralized learning baselines.
- 10 May 2025
- Published by Springer Nature
- p. 398-408
Video Moment Retrieval Based on Multimodal Information Fusion
- 10 May 2025
- Published by Springer Nature
- p. 398-408
Video Moment Retrieval Based on Multimodal Information Fusion
- 23 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 38 (6) , 300-308
Green and Economical Interaction in Metaverse: Challenges, Architecture, and Future Directions
This article presents an overview of existing surveys and reviews focusing on general and communication architectures (e.g., cloud- and P2P-based) within the metaverse. We propose GEIM, a green and economical multi-user interaction framework, to address inefficiencies in communication and computation, and to facilitate self-orchestrating and collaborative interactions. GEIM employs a flexible and intelligent communication and computation orchestration algorithm, along with caching strategy, to harmonize resources from remote cloud, edge cloud, and device layers for collaborative interactions. We deployed a GEIM prototype in real-world scenarios, demonstrating its capability to dynamically orchestrate interactive devices and computational resources, resulting in significant reductions in communication, computation, and memory overheads. Experimental results show that when GEIM collaborates with two edge nodes, it achieves a 17% reduction in GPU load, a 20% decrease in memory usage, and a minor 4% drop in frame rate compared to a single edge node setup. Last, we outline future research directions and conclude the research implications.
- 23 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Network
- Vol. 38 (6) , 300-308
Green and Economical Interaction in Metaverse: Challenges, Architecture, and Future Directions
This article presents an overview of existing surveys and reviews focusing on general and communication architectures (e.g., cloud- and P2P-based) within the metaverse. We propose GEIM, a green and economical multi-user interaction framework, to address inefficiencies in communication and computation, and to facilitate self-orchestrating and collaborative interactions. GEIM employs a flexible and intelligent communication and computation orchestration algorithm, along with caching strategy, to harmonize resources from remote cloud, edge cloud, and device layers for collaborative interactions. We deployed a GEIM prototype in real-world scenarios, demonstrating its capability to dynamically orchestrate interactive devices and computational resources, resulting in significant reductions in communication, computation, and memory overheads. Experimental results show that when GEIM collaborates with two edge nodes, it achieves a 17% reduction in GPU load, a 20% decrease in memory usage, and a minor 4% drop in frame rate compared to a single edge node setup. Last, we outline future research directions and conclude the research implications.
- 16 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 10 (6) , 2005-2018
LaMoSC: Large Language Model-Driven Semantic Communication System for Visual Transmission
The advancement of artificial intelligence (AI) has the potential to revolutionize network communication. The use of advanced feature extraction in semantic communication can enhance transmission capacity. However, relying solely on unimodal visual characteristics derived from images through these approaches may result in inaccuracies in decoding under low signal-to-noise ratio (SNR) conditions. This paper introduces LaMoSC, a semantic communication system driven by large language models (LLMs) that uses multimodal features to reconstruct raw visual information, thereby improving transmission quality. The system proposes an LLM-driven multimodal fusion semantic communication framework, which aims to expand unimodal transmission systems and enhance generalization ability. LaMoSC has designed an end-to-end encoding-decoding network that integrates visual and textual multimodal feature inputs. The design deeply integrates modal features using the attention mechanism. Comprehensive comparisons with state-of-the-art baselines across various datasets demonstrate the robustness of the proposed method, particularly highlighting its superiority in low SNR conditions. LaMoSC outperforms Deep-JSCC and multi-level semantic aware communication system (MLSC) by 5.5% and 2.6%, respectively, under low SNR conditions, such as 4 dB. Its exceptional generalization capacity sets it apart from other methods.
- 16 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 10 (6) , 2005-2018
LaMoSC: Large Language Model-Driven Semantic Communication System for Visual Transmission
The advancement of artificial intelligence (AI) has the potential to revolutionize network communication. The use of advanced feature extraction in semantic communication can enhance transmission capacity. However, relying solely on unimodal visual characteristics derived from images through these approaches may result in inaccuracies in decoding under low signal-to-noise ratio (SNR) conditions. This paper introduces LaMoSC, a semantic communication system driven by large language models (LLMs) that uses multimodal features to reconstruct raw visual information, thereby improving transmission quality. The system proposes an LLM-driven multimodal fusion semantic communication framework, which aims to expand unimodal transmission systems and enhance generalization ability. LaMoSC has designed an end-to-end encoding-decoding network that integrates visual and textual multimodal feature inputs. The design deeply integrates modal features using the attention mechanism. Comprehensive comparisons with state-of-the-art baselines across various datasets demonstrate the robustness of the proposed method, particularly highlighting its superiority in low SNR conditions. LaMoSC outperforms Deep-JSCC and multi-level semantic aware communication system (MLSC) by 5.5% and 2.6%, respectively, under low SNR conditions, such as 4 dB. Its exceptional generalization capacity sets it apart from other methods.
- 25 April 2024
- Published by Springer Nature
- p. 70-82
KiProL: A Knowledge-Injected Prompt Learning Framework for Language Generation
- 25 April 2024
- Published by Springer Nature
- p. 70-82
KiProL: A Knowledge-Injected Prompt Learning Framework for Language Generation
- 7 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 32 (23299290) , 853-867
FluGCF: A Fluent Dialogue Generation Model With Coherent Concept Entity Flow
The integration of external knowledge graphs into dialogue systems effectively mitigates the generation of generic and uninteresting responses. This approach, particularly the explicit modeling of conversation flows from related concept entities, facilitates the generation of semantically rich and informative responses. However, recent models guided by concept entity flows present two primary limitations: (1) a limited semantic understanding of the post message, which complicates the selection of highly relevant 1-hop concept entities, and (2) an inability to extract dynamic and diverse semantic relations between the post message and 2-hop concept entities. To address these issues, we introduce FluGCF, a novel model that fluently generates dialogues with coherent guidance from concept entity flows. FluGCF employs a ternary fusion to explicitly model multi-hop concept entity flows using a post-aware knowledge encoding mechanism. This mechanism learns semantic concept entity features from both word and sentence-level text features. Additionally, we design a corresponding ternary decoding mechanism that dynamically selects concept entities or words from the vocabulary to enhance fluency and diversity in dialogue generation. FluGCF, implemented in PyTorch, was extensively evaluated on a large-scale dataset, revealing that it surpasses baseline models, including the state-of-the-art knowledge-aware model ConceptFlow, by nearly 15% in terms of fluency. Furthermore, it demonstrated notable enhancements in coherence, diversity and informativeness.
- 7 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Audio, Speech, and Language Processing
- Vol. 32 (23299290) , 853-867
FluGCF: A Fluent Dialogue Generation Model With Coherent Concept Entity Flow
The integration of external knowledge graphs into dialogue systems effectively mitigates the generation of generic and uninteresting responses. This approach, particularly the explicit modeling of conversation flows from related concept entities, facilitates the generation of semantically rich and informative responses. However, recent models guided by concept entity flows present two primary limitations: (1) a limited semantic understanding of the post message, which complicates the selection of highly relevant 1-hop concept entities, and (2) an inability to extract dynamic and diverse semantic relations between the post message and 2-hop concept entities. To address these issues, we introduce FluGCF, a novel model that fluently generates dialogues with coherent guidance from concept entity flows. FluGCF employs a ternary fusion to explicitly model multi-hop concept entity flows using a post-aware knowledge encoding mechanism. This mechanism learns semantic concept entity features from both word and sentence-level text features. Additionally, we design a corresponding ternary decoding mechanism that dynamically selects concept entities or words from the vocabulary to enhance fluency and diversity in dialogue generation. FluGCF, implemented in PyTorch, was extensively evaluated on a large-scale dataset, revealing that it surpasses baseline models, including the state-of-the-art knowledge-aware model ConceptFlow, by nearly 15% in terms of fluency. Furthermore, it demonstrated notable enhancements in coherence, diversity and informativeness.
- Open Access
- 1 January 2023
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 2023 (1)
Beyond Words: An Intelligent Human‐Machine Dialogue System with Multimodal Generation and Emotional Comprehension
Intelligent service robots have become an indispensable aspect of modern‐day society, playing a crucial role in various domains ranging from healthcare to hospitality. Among these robotic systems, human‐machine dialogue systems are particularly noteworthy as they deliver both auditory and visual services to users, effectively bridging the communication gap between humans and machines. Despite their utility, the majority of existing approaches to these systems primarily concentrate on augmenting the logical coherence of the system’s responses, inadvertently neglecting the significance of user emotions in shaping a comprehensive communication experience. To tackle this shortcoming, we propose the development of an innovative human‐machine dialogue system that is both intelligent and emotionally sensitive, employing multimodal generation techniques. This system is architecturally comprised of three components: (1) data collection and processing, responsible for gathering and preparing relevant information, (2) a dialogue engine, which generates contextually appropriate responses, and (3) an interaction module, responsible for facilitating the communication interface between users and the system. To validate our proposed approach, we have constructed a prototype system and conducted an evaluation of the performance of the core dialogue engine by utilizing an open dataset. The results of our study indicate that our system demonstrates a remarkable level of multimodal generation response, ultimately offering a more human‐like dialogue experience.
- 1 January 2023
- Published by Wiley in International Journal of Intelligent Systems
- Vol. 2023 (1)
Open AccessBeyond Words: An Intelligent Human‐Machine Dialogue System with Multimodal Generation and Emotional Comprehension
Intelligent service robots have become an indispensable aspect of modern‐day society, playing a crucial role in various domains ranging from healthcare to hospitality. Among these robotic systems, human‐machine dialogue systems are particularly noteworthy as they deliver both auditory and visual services to users, effectively bridging the communication gap between humans and machines. Despite their utility, the majority of existing approaches to these systems primarily concentrate on augmenting the logical coherence of the system’s responses, inadvertently neglecting the significance of user emotions in shaping a comprehensive communication experience. To tackle this shortcoming, we propose the development of an innovative human‐machine dialogue system that is both intelligent and emotionally sensitive, employing multimodal generation techniques. This system is architecturally comprised of three components: (1) data collection and processing, responsible for gathering and preparing relevant information, (2) a dialogue engine, which generates contextually appropriate responses, and (3) an interaction module, responsible for facilitating the communication interface between users and the system. To validate our proposed approach, we have constructed a prototype system and conducted an evaluation of the performance of the core dialogue engine by utilizing an open dataset. The results of our study indicate that our system demonstrates a remarkable level of multimodal generation response, ultimately offering a more human‐like dialogue experience.