Bio

I am a Ph.D. student in Computer Science at New York University, advised by Prof. Qiaoyu Tan. My current research focuses on agentic reasoning, graph learning, and multimodal large language models, with an emphasis on building reasoning systems for structured data, efficient search-augmented agents, and multimodal graph learning.

Previously, I worked on high-speed networked systems at NYU under the supervision of Prof. Grace Liu and Prof. H. Jonathan Chao. I received my M.S. in Computer Engineering from NYU and my B.Eng. in Software Engineering from Northeastern University (China).

I'm open to opportunities. Feel free to get in touch via email.

News

  • [05/2026] One paper accepted to KDD 2026: MLaGA.
  • [05/2026] Started as a Research Scientist Intern at Adobe Research, San Jose.
  • [04/2026] One paper accepted to ACL 2026: AgentGL.
  • [03/2026] Two papers accepted to CVPR 2026: GraphVLM and Mario.
  • [02/2025] Joined the DIR Lab and began research on graph learning and agentic reasoning.

Publications

  • GraphSearch: Agentic Search-Augmented Reasoning for Zero-Shot Graph Learning [link]
    Jiajin Liu, Yuanfu Sun, Dongzhe Fan, Qiaoyu Tan
    arXiv preprint, 2026.
  • MLaGA: Multimodal Large Language and Graph Assistant [link]
    Dongzhe Fan, Jiajin Liu, Yi Fang, Djellel Difallah, Qiaoyu Tan
    Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2026. (Oral)
  • AgentGL: Towards Agentic Graph Learning with LLMs via Reinforcement Learning [link]
    Yuanfu Sun*, Kang Li*, Dongzhe Fan, Jiajin Liu, Qiaoyu Tan
    Annual Meeting of the Association for Computational Linguistics (ACL), 2026.
  • GraphVLM: Benchmarking Vision Language Models for Multimodal Graph Learning [Arxiv][Website]
    Jiajin Liu, Dongzhe Fan, Chuanhao Ji, Daochen Zha, Qiaoyu Tan
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026.
  • Mario: Multimodal Graph Reasoning with Large Language Models [link]
    Yuanfu Sun, Kang Li, Pengkang Guo, Jiajin Liu, Qiaoyu Tan
    IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026.

Systems Publications:

  • Heimdall: Towards Risk-Aware Network Management Outsourcing [link]
    Yuejie Wang, Qiutong Men, Yongting Chen, Jiajin Liu, Gengyu Chen, Ying Zhang, Guyue Liu, Vyas Sekar
    Network and Distributed System Security Symposium (NDSS), 2025.
  • Sifter: An Inversion-Free and Large-Capacity Programmable Packet Scheduler [link]
    Peixuan Gao, Anthony Dalleggio, Jiajin Liu, Chen Peng, Yang Xu, H. Jonathan Chao
    USENIX Symposium on Networked Systems Design and Implementation (NSDI), 2024.
  • Inversion impact of approximate PIFO to Start-Time Fair Queueing [link]
    Junda Song, Jiajin Liu, Peixuan Gao, Guyue Liu, H. Jonathan Chao
    Computer Networks, 2024.