Elamur Research⌕
☰
home / papers / 2026 / Mango: Multi-Agent Web Navigation via Global-View Optimization

Mango: Multi-Agent Web Navigation via Global-View Optimization

Paper · 2026 · Weixi Tong, Yifeng Di and Tianyi Zhang

Summary

LLM-based web agents usually begin at a site's homepage and descend through its hierarchy. On large, deeply nested websites this wastes the step budget on irrelevant branches and generic pages. MANGO addresses this by first building an approximate global map of the target site. It runs a bounded BFS crawl restricted to the same domain and ranks pages against the query with BM25. It also uses LLM-suggested keywords with Google's site: operator to reach pages the crawl misses. The highest-ranked pages become candidate entry points. The system treats choosing among them as a multi-armed bandit, using Thompson Sampling with Beta priors seeded from BM25 relevance. After each attempt, a reflection agent scores the trajectory, updates the posterior, and can retire dead-end URLs. An episodic memory keeps the agent from repeating failed actions. Across five LLM backbones on WebVoyager and WebWalkerQA, MANGO achieves higher success rates than AgentOccam and WebWalker. This shows that choosing good starting points matters as much as step-level decision-making.

Key points

  • With GPT-5-mini, MANGO reaches 63.6% success rate (SR) on WebVoyager, 7.3 points above the best baseline. On WebWalkerQA it reaches 52.5%, a 26.8-point gain.
  • Across all five backbones, absolute SR gains range from 3.1–7.3 points on WebVoyager and 4.6–26.8 points on WebWalkerQA. Action count is similar or lower with Qwen3 models but higher with GPT-5-mini.
  • Ablations show MANGO beats variants using random URL choice, search-based URL choice, and MCTS-guided navigation. The margin over the MCTS variant is 5.2–17.1 points.
  • The authors manually analyzed 323 WebWalkerQA failures. Running out of budget accounts for 52.4%, wrong sub-page navigation 24.6%, and reasoning mistakes 15.4%. Outdated gold answers cause 5.6% and reflection errors 2.0%.
  • Data and code are released at github.com/VichyTong/Mango.

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

🎓Purdue University West LafayetteWest Lafayette · United States23 works · 1,417 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksInformation Systems · 1,437 worksDecision Sciences · 209 worksManagement Science and Operations Research · 108 works

Cite

Cite as Weixi Tong, Yifeng Di and Tianyi Zhang. 2026. Mango: Multi-Agent Web Navigation via Global-View Optimization. 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.646

@inproceedings{tong2026mango,
  title = {{Mango: Multi-Agent Web Navigation via Global-View Optimization}},
  author = {Weixi Tong, Yifeng Di and Tianyi Zhang},
  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.646},
  url = {https://doi.org/10.18653/v1/2026.acl-long.646}
}

References

29 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Tamer Abuelsaad, Deepak Akkil, Prasenjit Dey, Ashish Jagmohan, Aditya Vempaty, and Ravi Kokku. 2024. Agent-e: From autonomous web navigation to foundational design principles in agentic systems. arXiv preprint arXiv:2407.13032. — [arXiv:2407.13032](https://arxiv.org/abs/2407.13032) find
  2. Saaket Agashe, Jiuzhou Han, Shuyu Gan, Jiachen Yang, Ang Li, and Xin Eric Wang. 2025. Agent s: An open agentic framework that uses computers like a human. In The Thirteenth International Conference on Learning Representations. find
  3. Deepak Akkil, Tamer Abuelsaad, Prasenjit Dey, Ashish Jagmohan, Aditya Vempaty, and Ravi Kokku. 2024. Agent-e: From autonomous web navigation to foundational design principles in agentic systems. In NeurIPS 2024 Workshop on Open-World Agents. find
  4. Antonis Antoniades, Albert Örwall, Kexun Zhang, Yuxi Xie, Anirudh Goyal, and William Yang Wang. 2025. SWE-search: Enhancing software agents with monte carlo tree search and iterative refinement. In The Thirteenth International Conference on Learning Representations. find
  5. Deepayan Chakrabarti, Ravi Kumar, Filip Radlinski, and Eli Upfal. 2008. Mortal multi-armed bandits. In Advances in Neural Information Processing Systems, volume 21. Curran Associates, Inc. find
  6. Xiang Deng, Yu Gu, Boyuan Zheng, Shijie Chen, Samuel Stevens, Boshi Wang, Huan Sun, and Yu Su. 2023. Mind2web: Towards a generalist agent for the web. In Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track.
  7. Bingzheng Gan, Yufan Zhao, Tianyi Zhang, Jing Huang, Li Yusu, Shu Xian Teo, Changwang Zhang, and Wei Shi. 2025. MASTER: A multi-agent system with LLM specialized MCTS. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human… find
  8. Yanchu Guan, Dong Wang, Zhixuan Chu, Shiyu Wang, Feiyue Ni, Ruihua Song, and Chenyi Zhuang. 2024. Intelligent agents with llm-based process automation. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD ’24, page 5018–5027, New York, NY, USA. Association for… find
  9. Izzeddin Gur, Hiroki Furuta, Austin V Huang, Mustafa Safdari, Yutaka Matsuo, Douglas Eck, and Aleksandra Faust. 2024. A real-world webagent with planning, long context understanding, and program synthesis. In The Twelfth International Conference on Learning Representations. find
  10. Hongliang He, Wenlin Yao, Kaixin Ma, Wenhao Yu, Yong Dai, Hongming Zhang, Zhenzhong Lan, and Dong Yu. 2024. WebVoyager: Building an end-to-end web agent with large multimodal models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), p…
  11. Jaekyeom Kim, Dong-Ki Kim, Lajanugen Logeswaran, Sungryull Sohn, and Honglak Lee. 2024. Auto-intent: Automated intent discovery and self-exploration for large language model web agents. find
  12. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 16531–16541, Miami, Florida, USA. Association for Computational Linguistics. find
  13. Jing Yu Koh, Stephen Marcus McAleer, Daniel Fried, and Ruslan Salakhutdinov. 2025. Tree search for language model agents. Transactions on Machine Learning Research. find
  14. Hanyu Lai, Xiao Liu, Iat Long Iong, Shuntian Yao, Yuxuan Chen, Pengbo Shen, Hao Yu, Hanchen Zhang, Xiaohan Zhang, Yuxiao Dong, and Jie Tang. 2024. Autowebglm: A large language model-based web navigating agent. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, K…
  15. Tao Li, Gang Li, Zhiwei Deng, Bryan Wang, and Yang Li. 2023. A zero-shot language agent for computer control with structured reflection. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 11261–11274, Singapore. Association for Computational Linguistics. find
  16. Kevin Qinghong Lin, Linjie Li, Difei Gao, Zhengyuan Yang, Shiwei Wu, Zechen Bai, Stan Weixian Lei, Lijuan Wang, and Mike Zheng Shou. 2025. Showui: One vision-language-action model for gui visual agent. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 19498– 19508. find
  17. Liangbo Ning, Ziran Liang, Zhuohang Jiang, Haohao Qu, Yujuan Ding, Wenqi Fan, Xiao-yong Wei, Shanru Lin, Hui Liu, Philip S Yu, and 1 others. 2025. A survey of webagents: Towards next-generation ai agents for web automation with large foundation models. In Proceedings of the 31st ACM SIGKDD Conferenc… find
  18. Stephen Robertson, Hugo Zaragoza, and 1 others. 2009. The probabilistic relevance framework: Bm25 and beyond. Foundations and Trends® in Information Retrieval, 3(4):333–389. find
  19. Weihao Tan, Wentao Zhang, Xinrun Xu, Haochong Xia, Ziluo Ding, Boyu Li, Bohan Zhou, Junpeng Yue, Jiechuan Jiang, Yewen Li, Ruyi An, Molei Qin, Chuqiao Zong, Longtao Zheng, YuJie Wu, Xiaoqiang Chai, Yifei Bi, Tianbao Xie, Pengjie Gu, and 9 others. 2025. Cradle: Empowering foundation agents towards ge… find
  20. Guanzhi Wang, Yuqi Zhang, Shuai Zhang, and 1 others. 2023. Voyager: An open-ended embodied agent with large language models. arXiv preprint arXiv:2305.16291. find
  21. Jialong Wu, Wenbiao Yin, Yong Jiang, Zhenglin Wang, Zekun Xi, Runnan Fang, Linhai Zhang, Yulan He, Deyu Zhou, Pengjun Xie, and Fei Huang. 2025. Web-Walker: Benchmarking LLMs in web traversal. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long P… find
  22. Wujiang Xu, Zujie Liang, Kai Mei, Hang Gao, Juntao Tan, and Yongfeng Zhang. 2025. A-mem: Agentic memory for LLM agents. In The Thirty-ninth Annual Conference on Neural Information Processing Systems.
  23. Ke Yang, Yao Liu, Sapana Chaudhary, Rasool Fakoor, Pratik Chaudhari, George Karypis, and Huzefa Rangwala. 2025. Agentoccam: A simple yet strong baseline for LLM-based web agents. In The Thirteenth International Conference on Learning Representations. find
  24. Xiao Yu, Baolin Peng, Vineeth Vajipey, Hao Cheng, Michel Galley, Jianfeng Gao, and Zhou Yu. 2025. Ex-ACT: Teaching AI agents to explore with reflective-MCTS and exploratory learning. In The Thirteenth International Conference on Learning Representations. find
  25. Yao Zhang, Zijian Ma, Yunpu Ma, and Volker Tresp. 2024. WebPilot: A Versatile and Autonomous Multi-Agent System for Web Task Execution with Strategic Exploration. In Proceedings of the AAAI Conference on Artificial Intelligence. find
  26. Longtao Zheng, Rundong Wang, Xinrun Wang, and Bo An. 2024. Synapse: Trajectory-as-exemplar prompting with memory for computer control. In The Twelfth International Conference on Learning Representations. find
  27. Andy Zhou, Kai Yan, Michal Shlapentokh-Rothman, Haohan Wang, and Yu-Xiong Wang. 2024a. Language agent tree search unifies reasoning acting and planning in language models. find
  28. Andy Zhou, Kai Yan, Michal Shlapentokh-Rothman, Haohan Wang, and Yu-Xiong Wang. 2024b. Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models. In International Conference on Machine Learning. find
  29. Shuyan Zhou, Frank F. Xu, Hao Zhu, Xuhui Zhou, Robert Lo, Abishek Sridhar, Xianyi Cheng, Tianyue Ou, Yonatan Bisk, Daniel Fried, Uri Alon, and Graham Neubig. 2024c. Webarena: A realistic web environment for building autonomous agents. In The Twelfth International Conference on Learning Representatio…

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/tong-2026-mango-multi-agent-web-navigation/