Elamur Research⌕
☰
home / papers / 2026 / AgentGL: Towards Agentic Graph Learning with LLMs via Reinforcement Learning

AgentGL: Towards Agentic Graph Learning with LLMs via Reinforcement Learning

Paper · 2026 · Yuanfu Sun, Kang Li, Dongzhe Fan et al.

Summary

Agentic LLM systems usually fetch evidence as plain text and ignore graph structure. This matters for Text-Attributed Graphs (TAGs) such as citation networks or product co-purchase graphs, where meaning comes from both node text and topology. The authors propose Agentic Graph Learning (AGL), in which an LLM agent alternates between exploring the graph and reasoning about what it has found. Their framework, AgentGL, is trained with reinforcement learning and gives the agent four graph-native retrieval tools that work at local and global scales.

Training has two stages. The first teaches the agent basic navigation using outcome-based rewards. The second makes searching more economical, so the agent reasons more before issuing new queries. A graph-conditioned curriculum, which gradually increases exploration difficulty, keeps both stages stable without step-level supervision.

Evaluated on several TAG benchmarks with different LLM backbones, AgentGL outperforms GraphLLM and GraphRAG baselines on node classification and link prediction. The results suggest that RL-trained, structure-aware agents are a promising alternative to static graph prompting.

Key points

  • Tools. The agent has four retrieval tools:
    • 1-hop and 2-hop neighborhood search, which favors shared neighbors and ranks candidates by a mix of query and target embeddings controlled by λ_r;
    • structure salience search based on precomputed PPR scores;
    • dense semantic search.
  • Main results. Absolute gains reach up to 17.5% on node classification and up to 28.4% on link prediction. Benchmarks include OGB-Arxiv, PubMed, Amazon-Computers and Amazon-Photo. The policy is optimized with GRPO and REINFORCE++.
  • Efficiency stage. This stage combines a Retrospective Termination Trigger (RTT) with Cognitive Density Regularization (CDR). Together they cut search cost by about 22% and raise accuracy by roughly 3%. Compared with the bootstrapping stage alone, the full pipeline uses about 17.5% fewer tool calls and gains 2.4% on node classification.
  • Ablations.
    • Without the coverage reward r_COV, the agent eventually stops searching altogether.
    • The best setting is λ_r = 0.5, which balances structural and semantic signals.
    • The curriculum (GCCL) steadies training and adds about 0.65% accuracy.
  • Limitations and code. The method covers only text-attributed graphs, not multimodal ones, and is sensitive to how training data is split between the two stages. Code is available at github.com/sunyuanfu/AgentGL.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓New York UniversityNew York · United States201 works · 100,180 citations🎓Tsinghua UniversityBeijing · China333 works · 87,019 citations🎓New York University ShanghaiShanghai · China8 works · 523 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksComputer Vision and Pattern Recognition · 2,869 works

Cite

Cite as Yuanfu Sun, Kang Li, Dongzhe Fan, Jiajin Liu and Qiaoyu Tan. 2026. AgentGL: Towards Agentic Graph Learning with LLMs via Reinforcement Learning. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.1161

@inproceedings{sun2026agentgl,
  title = {{AgentGL: Towards Agentic Graph Learning with LLMs via Reinforcement Learning}},
  author = {Yuanfu Sun, Kang Li, Dongzhe Fan, Jiajin Liu and Qiaoyu Tan},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1161},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1161}
}

References

55 references, 15 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, and 1 others. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774.
  2. Mingyang Chen, Linzhuang Sun, Tianpeng Li, Haoze Sun, Yijie Zhou, Chenzheng Zhu, Haofen Wang, Jeff Z Pan, Wen Zhang, Huajun Chen, and 1 others. 2025. Learning to reason with search for llms via reinforcement learning. arXiv preprint arXiv:2503.19470. find
  3. Runjin Chen, Tong Zhao, Ajay Jaiswal, Neil Shah, and Zhangyang Wang. 2024a. Llaga: Large language and graph assistant. arXiv preprint arXiv:2402.08170. — [arXiv:2402.08170](https://arxiv.org/abs/2402.08170) find
  4. Runjin Chen, Tong Zhao, AJAY KUMAR JAISWAL, Neil Shah, and Zhangyang Wang. 2024b. Llaga: Large language and graph assistant. In Forty-first International Conference on Machine Learning. find
  5. Zhikai Chen, Haitao Mao, Hang Li, Wei Jin, Hongzhi Wen, Xiaochi Wei, Shuaiqiang Wang, Dawei Yin, Wenqi Fan, Hui Liu, and 1 others. 2024c. Exploring the potential of large language models (llms) in learning on graphs. ACM SIGKDD Explorations Newsletter, 25(2):42–61. find
  6. Junnan Dong, Siyu An, Yifei Yu, Qian-Wen Zhang, Linhao Luo, Xiao Huang, Yunsheng Wu, Di Yin, and Xing Sun. 2025. Youtu-graphrag: Vertically unified agents for graph retrieval-augmented complex reasoning. arXiv preprint arXiv:2508.19855. — [arXiv:2508.19855](https://arxiv.org/abs/2508.19855) find
  7. Yi Fang, Dongzhe Fan, Daochen Zha, and Qiaoyu Tan. 2024. Gaugllm: Improving graph contrastive learning for text-attributed graphs with large language models. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 747–758. find
  8. Yunfan Gao, Yun Xiong, Xinyu Gao, Kangxiang Jia, Jinliu Pan, Yuxi Bi, Yixin Dai, Jiawei Sun, Haofen Wang, and Haofen Wang. 2023. Retrieval-augmented generation for large language models: A survey. arXiv preprint arXiv:2312.10997, 2(1).
  9. Jiayan Guo, Lun Du, Hengyu Liu, Mengyu Zhou, Xinyi He, and Shi Han. 2023. Gpt4graph: Can large language models understand graph structured data? an empirical evaluation and benchmarking. arXiv preprint arXiv:2305.15066. find
  10. Bernal Jiménez Gutiérrez, Yiheng Shu, Weijian Qi, Sizhe Zhou, and Yu Su. 2025. From rag to memory: Non-parametric continual learning for large language models. arXiv preprint arXiv:2502.14802. find
  11. Will Hamilton, Zhitao Ying, and Jure Leskovec. 2017. Inductive representation learning on large graphs. Advances in neural information processing systems, 30.
  12. Haoyu Han, Yu Wang, Harry Shomer, Kai Guo, Jiayuan Ding, Yongjia Lei, Mahantesh Halappanavar, Ryan A Rossi, Subhabrata Mukherjee, Xianfeng Tang, and 1 others. 2024. Retrieval-augmented generation with graphs (graphrag). arXiv preprint arXiv:2501.00309. find
  13. Xiaoxin He, Xavier Bresson, Thomas Laurent, Adam Perold, Yann LeCun, and Bryan Hooi. Harnessing explanations: Llm-to-lm interpreter for enhanced text-attributed graph representation learning. In The Twelfth International Conference on Learning Representations. find
  14. Xiaoxin He, Yijun Tian, Yifei Sun, Nitesh Chawla, Thomas Laurent, Yann LeCun, Xavier Bresson, and Bryan Hooi. 2024. G-retriever: Retrieval-augmented generation for textual graph understanding and question answering. Advances in Neural Information Processing Systems, 37:132876–132907.
  15. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. 2021. Measuring mathematical problem solving with the math dataset. arXiv preprint arXiv:2103.03874.
  16. Jian Hu, Jason Klein Liu, Haotian Xu, and Wei Shen. 2025. Reinforce++: Stabilizing critic-free policy optimization with global advantage normalization. Preprint, arXiv:2501.03262. find
  17. Weihua Hu, Matthias Fey, Marinka Zitnik, Yuxiao Dong, Hongyu Ren, Bowen Liu, Michele Catasta, and Jure Leskovec. 2020. Open graph benchmark: Datasets for machine learning on graphs. Advances in neural information processing systems, 33:22118–22133. find
  18. Jin Huang, Xingjian Zhang, Qiaozhu Mei, and Jiaqi Ma. 2023. Can llms effectively leverage graph structural information: when and why. arXiv preprint arXiv:2309.16595. — [arXiv:2309.16595](https://arxiv.org/abs/2309.16595) find
  19. Glen Jeh and Jennifer Widom. 2003. Scaling personalized web search. In Proceedings of the 12th international conference on World Wide Web, pages 271–279. find
  20. Bernal Jimenez Gutierrez, Yiheng Shu, Yu Gu, Michihiro Yasunaga, and Yu Su. 2024. Hipporag: Neurobiologically inspired long-term memory for large language models. Advances in Neural Information Processing Systems, 37:59532–59569.
  21. Bowen Jin, Chulin Xie, Jiawei Zhang, Kashob Kumar Roy, Yu Zhang, Zheng Li, Ruirui Li, Xianfeng Tang, Suhang Wang, Yu Meng, and 1 others. 2024. Graph chain-of-thought: Augmenting large language models by reasoning on graphs. arXiv preprint arXiv:2404.07103.
  22. Bowen Jin, Hansi Zeng, Zhenrui Yue, Jinsung Yoon, Sercan Arik, Dong Wang, Hamed Zamani, and Jiawei Han. 2025. Search-r1: Training llms to reason and leverage search engines with reinforcement learning. arXiv preprint arXiv:2503.09516.
  23. Thomas N Kipf and Max Welling. 2016. Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907.
  24. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, and 1 others. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in neural information processing systems, 33:9459–…
  25. Guohao Li, Matthias Müller, Bernard Ghanem, and Vladlen Koltun. 2021. Training graph neural networks with 1000 layers. In International conference on machine learning, pages 6437–6449. PMLR. find
  26. Rui Li, Jiwei Li, Jiawei Han, and Guoyin Wang. 2024. Similarity-based neighbor selection for graph llms. arXiv preprint arXiv:2402.03720. — [arXiv:2402.03720](https://arxiv.org/abs/2402.03720) find
  27. Xiaoxi Li, Guanting Dong, Jiajie Jin, Yuyao Zhang, Yujia Zhou, Yutao Zhu, Peitian Zhang, and Zhicheng Dou. 2025. Search-o1: Agentic search-enhanced large reasoning models. arXiv preprint arXiv:2501.05366.
  28. Jiajin Liu, Yuanfu Sun, Dongzhe Fan, and Qiaoyu Tan. 2026. Graphsearch: Agentic search-augmented reasoning for zero-shot graph learning. arXiv preprint arXiv:2601.08621. — [arXiv:2601.08621](https://arxiv.org/abs/2601.08621) find
  29. Yuyan Liu, Sirui Ding, Sheng Zhou, Wenqi Fan, and Qiaoyu Tan. 2024a. Moleculargpt: Open large language model (llm) for few-shot molecular property prediction. arXiv preprint arXiv:2406.12950. — [arXiv:2406.12950](https://arxiv.org/abs/2406.12950) find
  30. Zheyuan Liu, Xiaoxin He, Yijun Tian, and Nitesh V Chawla. 2024b. Can we soft prompt llms for graph learning tasks? In Companion Proceedings of the ACM on Web Conference 2024, pages 481–484. find
  31. Haoran Luo, Haihong E, Guanting Chen, Qika Lin, Yikai Guo, Fangzhi Xu, Zemin Kuang, Meina Song, Xiaobao Wu, Yifan Zhu, and Luu Anh Tuan. 2025. Graph-r1: Towards agentic graphrag framework via end-to-end reinforcement learning. Preprint, arXiv:2507.21892. — [arXiv:2507.21892](https://arxiv.org/abs/2507.21892) find
  32. Haitao Mao, Juanhui Li, Harry Shomer, Bingheng Li, Wenqi Fan, Yao Ma, Tong Zhao, Neil Shah, and Jiliang Tang. 2023. Revisiting link prediction: A data perspective. arXiv preprint arXiv:2310.00793. — [arXiv:2310.00793](https://arxiv.org/abs/2310.00793) find
  33. Ofir Press, Muru Zhang, Sewon Min, Ludwig Schmidt, Noah A Smith, and Mike Lewis. 2023. Measuring and narrowing the compositionality gap in language models. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 5687–5711. find
  34. Qwen, :, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, and 25 others. 2025. Qwen2.5 technical report. Preprint, arXiv:2412.15115. find
  35. Prithviraj Sen, Galileo Namata, Mustafa Bilgic, Lise Getoor, Brian Galligher, and Tina Eliassi-Rad. 2008. Collective classification in network data. AI magazine, 29(3):93–93. find
  36. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, YK Li, Yang Wu, and 1 others. 2024. Deepseekmath: Pushing the limits of mathematical reasoning in open language models. arXiv preprint arXiv:2402.03300.
  37. Oleksandr Shchur, Maximilian Mumme, Aleksandar Bojchevski, and Stephan Günnemann. 2018. Pitfalls of graph neural network evaluation. arXiv preprint arXiv:1811.05868. — [arXiv:1811.05868](https://arxiv.org/abs/1811.05868) find
  38. Yucheng Shi, Qiaoyu Tan, Xuansheng Wu, Shaochen Zhong, Kaixiong Zhou, and Ninghao Liu. 2024. Retrieval-enhanced knowledge editing for multi-hop question answering in language models. arXiv preprint arXiv:2403.19631. — [arXiv:2403.19631](https://arxiv.org/abs/2403.19631) find
  39. Huatong Song, Jinhao Jiang, Yingqian Min, Jie Chen, Zhipeng Chen, Wayne Xin Zhao, Lei Fang, and Ji-Rong Wen. 2025. R1-searcher: Incentivizing the search capability in llms via reinforcement learning. arXiv preprint arXiv:2503.05592.
  40. Yuanfu Sun, Kang Li, Pengkang Guo, Jiajin Liu, and Qiaoyu Tan. 2026. Mario: Multimodal graph reasoning with large language models. arXiv preprint arXiv:2603.05181. — [arXiv:2603.05181](https://arxiv.org/abs/2603.05181) find
  41. Yuanfu Sun, Zhengnan Ma, Yi Fang, Jing Ma, and Qiaoyu Tan. 2025. Graphicl: Unlocking graph learning potential in llms through structured prompt design. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 2440–2459. find
  42. Jiabin Tang, Yuhao Yang, Wei Wei, Lei Shi, Lixin Su, Suqi Cheng, Dawei Yin, and Chao Huang. 2024. Graphgpt: Graph instruction tuning for large language models. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 491–500. find
  43. Xianfeng Tang, Huaxiu Yao, Yiwei Sun, Yiqi Wang, Jiliang Tang, Charu Aggarwal, Prasenjit Mitra, and Suhang Wang. 2020. Investigating and mitigating degree-related biases in graph convoltuional networks. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management, pa… find
  44. Petar Velickovic, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Lio, Yoshua Bengio, and 1 others. 2017. Graph attention networks. stat, 1050(20):10–48550. find
  45. Hao Yan, Chaozhuo Li, Ruosong Long, Chao Yan, Jianan Zhao, Wenwen Zhuang, Jun Yin, Peiyan Zhang, Weihao Han, Hao Sun, and 1 others. 2023. A comprehensive study on text-attributed graphs: Benchmarking and rethinking. Advances in Neural Information Processing Systems, 36:17238–17264. find
  46. Hao Yan, Chaozhuo Li, Jun Yin, Zhigang Yu, Weihao Han, Mingzheng Li, Zhengxin Zeng, Hao Sun, and Senzhang Wang. 2025. When graph meets multimodal: benchmarking and meditating on multimodal attributed graph learning. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Min… find
  47. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025. Qwen3 technical report. arXiv preprint arXiv:2505.09388.
  48. Zhilin Yang, William Cohen, and Ruslan Salakhudinov. 2016. Revisiting semi-supervised learning with graph embeddings. In International conference on machine learning, pages 40–48. PMLR. find
  49. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik R Narasimhan, and Yuan Cao. 2022. React: Synergizing reasoning and acting in language models. In The eleventh international conference on learning representations.
  50. Muhan Zhang and Yixin Chen. 2018. Link prediction based on graph neural networks. Advances in neural information processing systems, 31. find
  51. Qihai Zhang, Xinyue Sheng, Yuanfu Sun, and Qiaoyu Tan. 2025. Trustglm: Evaluating the robustness of graphllms against prompt, text, and structure attacks. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V. 2, pages 5912–5923. find
  52. Jianan Zhao, Le Zhuo, Yikang Shen, Meng Qu, Kai Liu, Michael Bronstein, Zhaocheng Zhu, and Jian Tang. 2023. Graphtext: Graph reasoning in text space. arXiv preprint arXiv:2310.01089. — [arXiv:2310.01089](https://arxiv.org/abs/2310.01089) find
  53. Jiong Zhu, Yujun Yan, Lingxiao Zhao, Mark Heimann, Leman Akoglu, and Danai Koutra. 2020. Beyond homophily in graph neural networks: Current limitations and effective designs. Advances in neural information processing systems, 33:7793–7804. find
  54. Luyao Zhuang, Shengyuan Chen, Yilin Xiao, Huachi Zhou, Yujing Zhang, Hao Chen, Qinggang Zhang, and Xiao Huang. 2025. Linearrag: Linear graph retrieval augmented generation on large-scale corpora. arXiv preprint arXiv:2510.10114. — [arXiv:2510.10114](https://arxiv.org/abs/2510.10114) find
  55. Keyulu Xu, Chengtao Li, Yonglong Tian, Tomohiro Sonobe, Ken-ichi Kawarabayashi, and Stefanie Jegelka. 2018. Representation learning on graphs with jumping knowledge networks. In International conference on machine learning, pages 5453–5462. pmlr. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/sun-2026-agentgl-agentic-graph-learning-llms/