Haolong Chen · 陈 浩泷
Ph.D. Student · LLM Agents & Domain-Specific Intelligence
About Me
I am currently pursuing my Ph.D. in Computer and Information Engineering at The Chinese University of Hong Kong, Shenzhen under the supervision of Prof. Guangxu Zhu and Prof. Tsung-Hui Chang. I received my B.Eng. in Software Engineering from Jinan University in 2023.
My research lies at the intersection of LLM agents and domain-specific intelligence. I am particularly interested in:
- Self-evolving agents: enabling agents to improve their knowledge and behavior through interaction and feedback.
- Agent memory systems: building adaptive, reliable mechanisms for storing and retrieving experience.
- AI for communication networks: applying foundation models and agentic reasoning to network diagnosis and optimization.
I am always open to research collaborations and discussions.
News
- Aug. 2026 — ERSkill: Evolving for Skill-Guided Adaptive Memory Retrieval is now available on arXiv.
- Aug. 2026 — Error-Aware Reverse Auction Mechanism for Large Language Model Routing is now available on arXiv.
- Jun. 2026 — 🎉 DK-Root was accepted by IEEE Transactions on Networking.
- Jun. 2026 — QoEReasoner is now available on arXiv.
- May 2026 — 🎉 STM3 was accepted by SIGKDD 2026.
- May 2026 — 🎉 AdaMeZO was accepted by ICML 2026.
- Apr. 2026 — Three Birds, One Stone is now available on arXiv.
- Mar. 2026 — 🎉 FeedSign was accepted by IEEE Transactions on Mobile Computing.
- Jun. 2025 — 🎉 An Overview of Domain-Specific Foundation Models was accepted by Science China Information Sciences.
Featured Publications
* Equal contribution.
Science China Information Sciences (CCF-A) · 2026
An Overview of Domain-Specific Foundation Models: Key Technologies, Applications and Challenges
Full Publications
Grouped by research theme
Agent Harness -- Memory, Self-Evolution, Routing, etc.
AuditWeaver: Weaving Structured Safety Memory into Instance-Grounded Checklists for Agent Risk Assessment
E3Mem: Evolving Evidence Connection from Task Traces for Long-Term Agent Memory
ERSkill: Evolving for Skill-Guided Adaptive Memory Retrieval
Error-Aware Reverse Auction Mechanism for Large Language Model Routing
LLM Post-Training
Three Birds, One Stone: Solving the Communication-Memory-Privacy Trilemma in LLM Fine-tuning Over Wireless Networks with Zeroth-Order Optimization
FeedSign: Robust and Communication-Efficient Federated Fine-tuning of Large Models for Edge AI
Domain-Specific Data Mining & Reasoning
From Root Causes to Fault Chains: Constrained Semantic Grounding for Agentic Diagnosis of RAN QoE Degradation
Learning to Detect and Diagnose Anomalies from Heterogeneous Supervision
QoEReasoner: An Agentic Reasoning Framework for Automated and Explainable QoE Diagnosis in RANs
SemiRoot: A Semi-Supervised Deep Learning Framework for Root-Cause Analysis of QoE Degradations in Mobile Networks
An Overview of Domain-Specific Foundation Models: Key Technologies, Applications and Challenges
First Token Probability Guided RAG for Telecom Question Answering
Domain-Specific Foundation-Model Customization: Theoretical Foundation and Key Technology
Experience
Meituan
Intern · Foundational R&D Platform · Mentor: Jianyu Wen
Research focus: agent guardrails / judges, and agent self-evolution.
Shenzhen Research Institute of Big Data
Research Assistant
Patents
-
基于大模型智能体的数据查询方法、装置、设备及介质
-
基于误差感知的多模型调度方法、装置、设备及存储介质
-
通信网络数据异常根因推理方法、装置、设备和存储介质
-
网络质差根因识别模型的训练方法、装置、设备及介质
-
无线通信中的流量预测方法、装置、电子设备及存储介质
-
基于多模态大模型的通信指标预测方法以及相关设备
-
模型训练方法、文本分类方法、装置、电子设备及介质
-
通信和内存高效的大模型分布式训练方法、文本分类方法
-
频谱效率的预测方法、装置、电子设备及存储介质
Education & Service
Ph.D. in Computer and Information Engineering
The Chinese University of Hong Kong, Shenzhen
Present
B.Eng. in Software Engineering
Jinan University
2023
Academic Service
Reviewer for ICLR, IEEE Transactions on Mobile Computing, ICASSP, WCNC, ICC, GLOBECOM, ICCC, PIMRC.