I am now a senior researcher at Tencent Youtu Lab, specializing in the research and development of Large Language Models and downstream directions. I obtained my Ph.D. degree from The Hong Kong Polytechnic University in 2025, Department of Computing, proudly advised by Prof. Xiao Huang and Prof. Qing Li. I have received with papers published on top-tier conferences and journals, e.g., NeurIPS, ICML, ICLR, ACL, EMNLP, SIGIR, KDD, WWW and TKDE, etc.

I am currently passionate about researching:
🥑 Large Language Models (LLMs): Diffusion Language Models (DLMs), Long-context Data Construction, SFT, RL
🥨 Multimodal Large Language Models (MLLMs): Native MLLMs, Interleaved Text-Image Data
🍤 Retrieval-Augmented Generation (RAG): GraphRAG, Agentic RAG, Multimodal RAG
🥓 Agents: Agentic Search, Agent Memory (Latent Reasoning, Compression)

We are actively recruiting ‘Qingyun’ Talent Program graduates and research interns to explore Memory, GraphRAG and related directions. Please feel free to drop me an email with your CV!

hanson[dot]{last_name}[at]connect[dot]polyu[dot]hk

📣 Highlights

[2026.05] 🔥 We release the first survey of Native Multimodal Modeling, Toward Native Multimodal Modeling: A Roadmap
[2026.01] 🎉 Our Youtu-GraphRAG has been accepted to ICLR 2026!
[2025.12] We release Youtu-LLM v1, a super effective pre-trained LLM for agents.
[2025.12] I will give a talk at CCF ChinaData (中国数据大会), GraphRAG Forum, in Guiyang City.
[2025.10] ✨ Our Youtu-GraphRAG has been admitted by the DeepSeek Official Awesome Integrations.
[2025.05] I passed the Ph.D. oral defense! Thanks to my distinguished panel members: Prof.Yuanqing Zheng(PolyU), Prof. Yangqiu Song (HKUST), Prof. Yixin Cao (FDU)
[2025.04] I will join Tencent Youtu Lab through `Qingyun’ Top Talent Program (腾讯青云计划).
[2025.03] I will give a talk at SCAI-SHUFE about KGQA with LLMs.
[2025.02] I am nominated by both the Department of Computing and the Faculty of Computer and Mathematical Science as the Best Research Postgraduate Student of 2024.
[2023.04] I passed the confirmation as a Ph.D. Candidate. Thanks to the distinguished panel members: Prof. Song Guo (HKUST) and Prof. Bo Li (PolyU).

🕹 From Research to Production

🚩[ICLR’ 26]Youtu-GraphRAG


🚩 Deep Tabular Research

🎙 Talks

CCF ChinaData 中国数据大会 - 🎈Graph+Agents

Tencent Youtu Lab - 🎈GraphRAG

SCAI-SHUFE - 🎈KGQA with LLMs

📻 Social Media

🎈Youtu-GraphRAG v1.0
机器之心, Tencent Cloud (腾讯云), Tencent Youtu Lab (腾讯优图实验室), 老刘说NLP
🎈GraphRAG-Bench
NebulaGraph Tech Community, Tencent Youtu Lab, 老刘说NLP

✍ Services

Program Committee Member
NeurIPS(2024-2025), ICML (2025-2026), ICLR (2025-2026), ACL (2025-2026), WWW (2024-2026), AAAI (2026-2027), WSDM 2027, NLPCC 2026

🏆 Awards

2026 Tencent Outstanding Contributor
2025 Best Research Postgraduate Student nominated by the Department of Computing in PolyU and Faculty of Computer and Mathematical Sciences.
2025 Tencent ‘Qingyun’ Top Talent Program（青云计划）
2025 Ant Group ‘Ant Star’ Top Talent Program（蚂蚁星）
2024 NeurIPS Scholar Award
2023 Hong Kong Polytechnic University Additional Top Conference Grant

🚀 Selected Publications (first-/corresponding author)

*for co-first author, † for corresponding author
1. Youtu-GraphRAG: Vertically Unified Agents for Graph Retrieval-Augmented Complex Reasoning.
Junnan Dong, Siyu An, Yifei Yu, Qian-wen Zhang, Linhao Luo, Xiao Huang, Yunsheng Wu, Di Yin, Xing Sun
ICLR 2026 (✨admitted by the DeepSeek Official Awesome Integrations Link)
2. Deep Tabular Research via Continual Experience-Driven Execution.
Junnan Dong, Chuang Zhou, Zheng Yuan, Yifei Yu, Siyu An, Di Yin, Xing Sun, Feiyue Huang
arXiv
3. Cost-efficient Knowledge-based Question Answering with Large Language Models.
Junnan Dong, Qinggang Zhang, Chuang Zhou, Hao Chen, Daochen Zha, Xiao Huang
NeurIPS 2024
4. Modality-Aware Integration with Large Language Models for Knowledge-based VQA.
Junnan Dong , Qinggang Zhang, Huachi Zhou, Daochen Zha, Pai Zheng, Xiao Huang
ACL Main 2024
5. Hierarchy-Aware Multi-Hop Question Answering over Knowledge Graphs.
Junnan Dong , Qinggang Zhang, Xiao Huang, Keyu Duan, Qiaoyu Tan, Zhimeng Jiang
WWW 2023
6. Active Ensemble Learning for Knowledge Graph Error Detection.
Junnan Dong, Qinggang Zhang, Xiao Huang, Qiaoyu Tan, Daochen Zha, Zihao Zhao
WSDM 2023
7. CLR-Bench: Evaluating Large Language Models in College-level Reasoning.
Junnan Dong, Zijin Hong, Yuanchen Bei, Feiran Huang, Xinrun Wang, Xiao Huang
arXiv
8. Gradual Study Advising with Course Knowledge Graphs.
Junnan Dong, Wentao Li, Yaowei Wang, Qing Li,
George Baciu, Jiannong Cao, Xiao Huang, Richard Li Chen, Peter Ng
ICWL 2023
9. KnowGPT: Knowledge Graph Prompting for Large Language Models.
Qinggang Zhang*, Junnan Dong* , Hao Chen, Xiao Huang, Daochen Zha, Zailiang Yu
NeurIPS 2024
10. GraphRAG-Bench: Challenging Domain-Specific Reasoning for Graph Retrieval-Augmented Generation.
Yilin Xiao*, Junnan Dong*†, Chuang Zhou, Su Dong, Qian-wen Zhang, Di Yin, Xing Sun, Xiao Huang
arXiv
11. HyperRAG: Query-centric Graph Retrieval-Augmented Generation with Hyperbolic Structuring.
Chuang Zhou, Junnan Dong (Project Lead), Yilin Xiao, Su Dong, Shengyuan Chen, Di Yin, Sun Xing, Xiao Huang
ACL Main 2026
12. Collision to Cognition: Hash-Driven Graph Construction for Efficient RAG.
Chuang Zhou, Junnan Dong†, Linhao Luo, Zhaozhuo Xu, Yilin Xiao, Junnan Dong, Siyu An, di yin, Xing Sun, Xiao Huang
ACL Main 2026
13. Disentangling Long-Term Memory via Latent Neuro-Symbolic Resaoning.
Cai Ke, Xinghao Chen, Xiaoyu Shen, Keyu Chen, Siyu An, Junnan Dong†, Ruifeng Xu, Ruizhi QIao, Xing Sun
arXiv
14. Toward Native Multimodal Modeling: A Roadmap.
Siyu An, Junru Lu, Junnan Dong*† et al.
EMNLP 2026 Findings