邱瑞鸿 /`ray hong chill/

I am currently an Assistant Professor and an ARC DECRA fellow (2025-2028) in School of Electrical Engineering and Computer Science (EECS) at The University of Queensland (UQ). Here is my short bio.

My current research mainly focuses on:

1. LLM post-training.
2. Agentic RL.
3. Diffusion LLM.

Previously, I worked on IR, RecSys, GNNs, and GFMs.

Recruitment

I am actively looking for (1-2) self-motivated PhD students in Year 2027, all fully funded!

• [For prospective PhD students] / [博士招生中文].
• [For master thesis, bachelor honours or Summer/Winter Research students at UQ].
• [For interns / visitors].

Recent News

• 10.2025 Our paper, “Beyond Static LLM Policies: Imitation-Enhanced Reinforcement Learning for Recommendation” is selected as Best Paper Finalist, ICDM 2025.

• 08.2024 ARC DECRA project funded, “Lifelong Paradigms for Versatile, Robust and Agile Recommender Systems” (2025-2028)

• 06.2024 Our team gives a talk, “Effective Representation Learning for Legal Case Retrieval”, at IR Seminar, the University of Glasgow. [slides]

• 05.2024 Our team gives a talk, “Effective Representation Learning for Legal Case Retrieval”, at THUIR, Tsinghua University. [slides]

• 03.2024 Give a talk, “Graph Learning Methods in Session-based Recommendations and Legal Case Retrieval”, at IRonGraphs Workshop at ECIR 2024. [slides]
• Past news

Selected Research

Google Scholar page includes the full publication list.

LLM Post-training

TRN-R1-Zero: Text-rich Network Reasoning via LLMs with Reinforcement Learning Only Yilun Liu, Ruihong Qiu, Zi Huang ACL 2026 (Main, Oral) arXiv / code We introduce TRN-R1-Zero, a RL-based LLM post-training method for effective reasoning over text-rich networks, such as citation, hyperlink, social and co-purchase domains.

Out-of-distribution on Graphs

	Text Meets Topology: Rethinking Out-of-distribution Detection in Text-Rich Networks Danny Wang, Ruihong Qiu, Guangdong Bai, Zi Huang EMNLP 2025 (Main) arXiv / code We introduce TextTopoOOD, a framework for modeling diverse OOD scenarios on text-rich networks, and propose TNT-OOD, a novel detection method that captures the intricate interplay between text and topology.
	GOLD: Graph Out-of-Distribution Detection via Implicit Adversarial Latent Generation Danny Wang, Ruihong Qiu, Guangdong Bai, Zi Huang ICLR 2025 (Spotlight) arXiv / OpenReview / code We propose the GOLD framework for graph OOD detection, an implicit adversarial learning pipeline with synthetic OOD exposure without pre-trained models.
	Does Homophily Help in Robust Test-time Node Classification? Yan Jiang, Ruihong Qiu, Zi Huang WSDM 2026 (Oral) arXiv / code We propose the GrapHoST framework for graph learning to conduct test-time graph transformation based on homophily to enhance the robustness of graph models.

Graph Condensation

	GCondenser: Benchmarking Graph Condensation Yilun Liu, Ruihong Qiu, Zi Huang CIKM 2025 arXiv / code We introduce a benchmark for graph condensation with a thorough methodology development method and an extensive evaluation protocol.
	PUMA: Efficient Continual Graph Learning with Graph Condensation Yilun Liu, Ruihong Qiu, Yanran Tang, Hongzhi Yin, Zi Huang TKDE 2024 arXiv / code We extend the Condense-and-Train (CaT) continual graph learning algorithm with a more efficient and effective, psudo-label guided memory bank (PUMA🐆) framework.
	CaT: Balanced Continual Graph Learning with Graph Condensation Yilun Liu, Ruihong Qiu, Zi Huang ICDM 2023 arXiv / code We introduce a Condense-and-Train (CaT🐱) memory-based continual graph learning algorithm using graph condensation to construct a more representative memory bank. And a Train-in-Memory continual learning scheme can further alleviate the imbalanced training issue in Class Incremental Learning.

Legal Case Retrieval

	CaseLink: Inductive Graph Learning for Legal Case Retrieval Yanran Tang, Ruihong Qiu, Hongzhi Yin, Xue Li, Zi Huang SIGIR 2024 arXiv / code We introduce an inductive graph learning paradigm for legal case retrieval to tackle the challenge of unseen testing query and candidate cases.
	CaseGNN++: Graph Contrastive Learning for Legal Case Retrieval with Graph Augmentation Yanran Tang, Ruihong Qiu, Yilun Liu, Xue Li, Zi Huang TOIS 2024 (under review) arXiv / code We introduce an extended CaseGNN++ method with graph augmentations based on the CaseGNN framework.
	CaseGNN: Graph Neural Networks for Legal Case Retrieval with Text-Attributed Graphs Yanran Tang, Ruihong Qiu, Yilun Liu, Xue Li, Zi Huang ECIR 2024 arXiv / code We introduce a structural modelling of law case for effective retrieval with the aid of summarisation from LLM and grpah neural networks.

Team

• Xiaoguang Qiao, UQ EECS PhD (1.2026-, co-advise with Helen Huang and Yujun Cai)
• Xiaoqian Hu, UQ EECS PhD (10.2025-, co-advise with Sen Wang)
• Chenke Xu, UQ EECS PhD (7.2025-, co-advise with Xue Li)
• Boyu Luo, UQ EECS PhD (7.2024-, co-advise with Helen Huang and Guangdong Bai)
• Yan Jiang, UQ EECS PhD (1.2024-, co-advise with Helen Huang and Guangdong Bai)
• Danny Wang, UQ EECS PhD (1.2024-, co-advise with Helen Huang and Guangdong Bai)
• Hrishikesh Patel, UQ EECS PhD (4.2023, co-advise with Sen Wang)
• Yilun Liu, UQ EECS PhD (1.2023-, co-advise with Helen Huang)

Alumni

• Yi Zhang, UQ EECS PhD -> SMU Postdoc (7.2024-12.2025, co-advise with Sen Wang and Jiajun Liu)
• Jingyu Ge, UQ ACWEB PhD -> UQ ACWEB Postdoc (1.2022-12.2025, co-advise with Zhiguo Yuan, Helen Huang, and Jiuling Li)

Service

• Conference organisation: Area Chair at NLPCC’25; PhD Symposium Co-Chair at WWW’25; Program Committee Co-Chair at ADC’23; PhD Forum Co-Chair at AJCAI’23

MISC

I speak Cantonese, Mandarin and English.

Updated on 22/04/2026.