Chuanfu Zhang
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
Research Subjects
Chuanfu Zhang has a total of 77 co-authors. Here are the top co-authors by number of shared publications:
- Tao Zhang
School of Systems Science and Engineering, Sun Yat-Sen University, Guangzhou, China
9 shared publications - Zong-Fu Luo
Sun Yat-sen University
6 shared publications - Gou Tan
School of Systems Science and Engineering, Sun Yat-Sen University, Guangzhou, China
4 shared publications - Pengfei Chen
School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China
4 shared publications - Zibin Zheng
Sun Yat-Sen University, Guangzhou, China
4 shared publications - Chuan Chen
School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou, China
3 shared publications
Latest Publications
- Open Access
- 18 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Software Engineering and Methodology
- Vol. 35 (12) , 1-33
LLMRCA: Multilevel Root Cause Analysis for LLM Applications Using Multimodal Observability Data
Ensuring the reliability of Large Language Model (LLM) applications in production environments is critical as LLMs are widely integrated into software systems. However, fault diagnosis in these applications is challenging due to distributed deployment and complex component interactions. Existing Root Cause Analysis (RCA) methods fall short when applied to LLM applications because they ignore two key challenges: unstable response times and response quality-related silent faults. To address these challenges, we propose LLMRCA, the first multilevel and unsupervised RCA framework that leverages multimodal observability data to identify root causes in LLM applications. LLMRCA constructs a heterogeneous causal graph from metrics, logs, and traces, and employs a Residual Graph Attention autoencoder to calculate reconstruction errors for RCA. LLMRCA also introduces request classification to handle unstable response times and verification to reduce false positives. Experiments on a retrieval-augmented generation LLM application show that LLMRCA outperforms eight baseline methods, achieving up to \(5.1\times\) more accurate results than baselines for performance problems and 92.86% top-1 ranking accuracy for response-quality problems. Furthermore, the ablation studies confirm the contributions of each data modality and different framework modules to the effectiveness. Generally, our framework provides a practical approach for diagnosing both performance and quality faults in LLM applications.- 18 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Software Engineering and Methodology
- Vol. 35 (12) , 1-33
Open AccessLLMRCA: Multilevel Root Cause Analysis for LLM Applications Using Multimodal Observability Data
Ensuring the reliability of Large Language Model (LLM) applications in production environments is critical as LLMs are widely integrated into software systems. However, fault diagnosis in these applications is challenging due to distributed deployment and complex component interactions. Existing Root Cause Analysis (RCA) methods fall short when applied to LLM applications because they ignore two key challenges: unstable response times and response quality-related silent faults. To address these challenges, we propose LLMRCA, the first multilevel and unsupervised RCA framework that leverages multimodal observability data to identify root causes in LLM applications. LLMRCA constructs a heterogeneous causal graph from metrics, logs, and traces, and employs a Residual Graph Attention autoencoder to calculate reconstruction errors for RCA. LLMRCA also introduces request classification to handle unstable response times and verification to reduce false positives. Experiments on a retrieval-augmented generation LLM application show that LLMRCA outperforms eight baseline methods, achieving up to \(5.1\times\) more accurate results than baselines for performance problems and 92.86% top-1 ranking accuracy for response-quality problems. Furthermore, the ablation studies confirm the contributions of each data modality and different framework modules to the effectiveness. Generally, our framework provides a practical approach for diagnosing both performance and quality faults in LLM applications. - Open Access
- 7 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 1780-1792
AgentChaos: Chaos Engineering for Agent Systems via Programmatic Fault Injection
- 7 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 1780-1792
Open AccessAgentChaos: Chaos Engineering for Agent Systems via Programmatic Fault Injection
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (7) , 2337-2350
TransformKV: Optimizing Multi-Turn Conversational Services in LLMs via KV Cache Transformation
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computers
- Vol. 75 (7) , 2337-2350
TransformKV: Optimizing Multi-Turn Conversational Services in LLMs via KV Cache Transformation
- 7 February 2026
- Published by Elsevier in Journal of Systems and Software
- Vol. 236
LogGen: Integrating traditional model and LLM with code analysis for precise log generation
- 7 February 2026
- Published by Elsevier in Journal of Systems and Software
- Vol. 236
LogGen: Integrating traditional model and LLM with code analysis for precise log generation
- Open Access
LIDL: LLM Integration Defect Localization via Knowledge Graph-Enhanced Multi-Agent Analysis
LLM-integrated software, which embeds or interacts with large language models (LLMs) as functional components, exhibits probabilistic and context-dependent behaviors that fundamentally differ from those of traditional software. This shift introduces a new category of integration defects that arise not only from code errors but also from misaligned interactions among LLM-specific artifacts, including prompts, API calls, configurations, and model outputs. However, existing defect localization techniques are ineffective at identifying these LLM-specific integration defects because they fail to capture cross-layer dependencies across heterogeneous artifacts, cannot exploit incomplete or misleading error traces, and lack semantic reasoning capabilities for identifying root causes. To address these challenges, we propose LIDL, a multi-agent framework for defect localization in LLM-integrated software. LIDL (1) constructs a code knowledge graph enriched with LLM-aware annotations that represent interaction boundaries across source code, prompts, and configuration files, (2) fuses three complementary sources of error evidence inferred by LLMs to surface candidate defect locations, and (3) applies context-aware validation that uses counterfactual reasoning to distinguish true root causes from propagated symptoms. We evaluate LIDL on 146 real-world defect instances collected from 105 GitHub repositories and 16 agent-based systems. The results show that LIDL significantly outperforms five state-of-the-art baselines across all metrics, achieving a Top-3 accuracy of 0.64 and a MAP of 0.48, which represents a 64.1% improvement over the best-performing baseline. Notably, LIDL achieves these gains while reducing cost by 92.5%, demonstrating both high accuracy and cost efficiency.
Open AccessLIDL: LLM Integration Defect Localization via Knowledge Graph-Enhanced Multi-Agent Analysis
LLM-integrated software, which embeds or interacts with large language models (LLMs) as functional components, exhibits probabilistic and context-dependent behaviors that fundamentally differ from those of traditional software. This shift introduces a new category of integration defects that arise not only from code errors but also from misaligned interactions among LLM-specific artifacts, including prompts, API calls, configurations, and model outputs. However, existing defect localization techniques are ineffective at identifying these LLM-specific integration defects because they fail to capture cross-layer dependencies across heterogeneous artifacts, cannot exploit incomplete or misleading error traces, and lack semantic reasoning capabilities for identifying root causes. To address these challenges, we propose LIDL, a multi-agent framework for defect localization in LLM-integrated software. LIDL (1) constructs a code knowledge graph enriched with LLM-aware annotations that represent interaction boundaries across source code, prompts, and configuration files, (2) fuses three complementary sources of error evidence inferred by LLMs to surface candidate defect locations, and (3) applies context-aware validation that uses counterfactual reasoning to distinguish true root causes from propagated symptoms. We evaluate LIDL on 146 real-world defect instances collected from 105 GitHub repositories and 16 agent-based systems. The results show that LIDL significantly outperforms five state-of-the-art baselines across all metrics, achieving a Top-3 accuracy of 0.64 and a MAP of 0.48, which represents a 64.1% improvement over the best-performing baseline. Notably, LIDL achieves these gains while reducing cost by 92.5%, demonstrating both high accuracy and cost efficiency.
- 11 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (9) , 9004-9019
Federated Domain-Independent Prototype Learning With Alignments of Representation and Parameter Spaces for Feature Shift
Federated learning provides a privacy-preserving modeling schema for distributed data, which coordinates multiple clients to collaboratively train a global model. However, data stored in different clients may be collected from diverse domains, and the resulting feature shift is prone to the degraded performance of global model. In this paper, we propose a Federated Domain-Independent Prototype Learning (FedDP) method with Alignments of Representation and Parameter Spaces for Feature Shift. Concretely, FedDP aims to eliminate the domain-specific information and explore the pure representations via information bottleneck, thus integrating the local and global domain-independent prototypes, respectively. To align the cross-domain representation spaces, the global domain-independent prototypes serve as the supervised signals to enable local intra-class representations to approach them. Further, to mitigate the divergences of optimization directions between multiple clients induced by the feature shift, the global representations are yielded by the global model on the client-side and guide the learning of local representations, thus unifying the parameter spaces of multiple local models. We derive the theoretical lower bound of the optimization objective based on mutual information, which is transformed into a computable loss. The proposed FedDP can be applied in the scenarios of homogeneous and heterogeneous models. Extensive experiments are conducted on three challenging multi-domain datasets. The experimental results illustrate the superiority of FedDP compared with state-of-the-art federated learning methods.
- 11 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 24 (9) , 9004-9019
Federated Domain-Independent Prototype Learning With Alignments of Representation and Parameter Spaces for Feature Shift
Federated learning provides a privacy-preserving modeling schema for distributed data, which coordinates multiple clients to collaboratively train a global model. However, data stored in different clients may be collected from diverse domains, and the resulting feature shift is prone to the degraded performance of global model. In this paper, we propose a Federated Domain-Independent Prototype Learning (FedDP) method with Alignments of Representation and Parameter Spaces for Feature Shift. Concretely, FedDP aims to eliminate the domain-specific information and explore the pure representations via information bottleneck, thus integrating the local and global domain-independent prototypes, respectively. To align the cross-domain representation spaces, the global domain-independent prototypes serve as the supervised signals to enable local intra-class representations to approach them. Further, to mitigate the divergences of optimization directions between multiple clients induced by the feature shift, the global representations are yielded by the global model on the client-side and guide the learning of local representations, thus unifying the parameter spaces of multiple local models. We derive the theoretical lower bound of the optimization objective based on mutual information, which is transformed into a computable loss. The proposed FedDP can be applied in the scenarios of homogeneous and heterogeneous models. Extensive experiments are conducted on three challenging multi-domain datasets. The experimental results illustrate the superiority of FedDP compared with state-of-the-art federated learning methods.
- 15 March 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 169
Delay-cost computation offloading for on-board emergency tasks in LEO Satellite Edge Computing networks
- 15 March 2025
- Published by Elsevier in Future Generation Computer Systems
- Vol. 169
Delay-cost computation offloading for on-board emergency tasks in LEO Satellite Edge Computing networks
- Open Access
- 21 February 2025
- Published by Association for Computing Machinery (ACM)
- p. 183-189
FANET-Sim: A Simulation Framework for UAV Swarm Communication
- 21 February 2025
- Published by Association for Computing Machinery (ACM)
- p. 183-189
Open AccessFANET-Sim: A Simulation Framework for UAV Swarm Communication
- 27 December 2024
- Published by Springer Nature
- p. 394-404
A Robust Connectivity Maintenance Method for Multi-UAVs Network Under Obstacle Environment
- 27 December 2024
- Published by Springer Nature
- p. 394-404
A Robust Connectivity Maintenance Method for Multi-UAVs Network Under Obstacle Environment