paper-with-me

Papers

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

2026-04-20 · Weixi Tong, Yifeng Di, Tianyi Zhang arxiv

Existing web agents typically initiate exploration from the root URL, which is inefficient for complex websites with deep hierarchical structures. Without a global view of the website's structure, agents frequently fall into navigation traps, explore irrelevant branches, or fail to reach target information within a limited budget. We propose Mango, a multi-agent web navigation method that leverages the website structure to dynamically determine optimal starting points. We formulate URL selection as a multi-armed bandit problem and employ Thompson Sampling to adaptively allocate the navigation budget across candidate URLs. Furthermore, we introduce an episodic memory component to store navigation history, enabling the agent to learn from previous attempts. Experiments on WebVoyager demonstrate that Mango achieves a success rate of 63.6% when using GPT-5-mini, outperforming the best baseline by 7.3%. Furthermore, on WebWalkerQA, Mango attains a 52.5% success rate, surpassing the best baseline by 26.8%. We also demonstrate the generalizability of Mango using both open-source and closed-source models as backbones. Our data and code are open-source and available at https://github.com/VichyTong/Mango.

📄 PDF Abstract BibTeX arXiv:2604.18779

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

MANGO: A Benchmark for Evaluating Mapping and Navigation Abilities of Large Language Models

2024-03-29 · Peng Ding, Jiading Fang, Peng Li, Kangrui Wang 외

Large language models such as ChatGPT and GPT-4 have recently achieved astonishing performance on a variety of natural language processing tasks. In this paper, we propose MANGO, a benchmark to evaluate their capabilitie…

Language ModelingLanguage ModellingLarge Language ModelQuestion Answering

Reinforced Collaboration in Multi-Agent Flow Networks

2026-05-13 · Zheng Wang, Yuang Liu, Yangkai Ding arxiv

Multi-agent systems provide a powerful way to extend large language models (LLMs) by decomposing a complex task into specialized subtasks handled by different agents. However, their performance is often hindered by error…

Reinforcement Learning

Constructing coherent spatial memory in LLM agents through graph rectification

2025-10-05 · Puzhen Zhang, Xuyang Chen, Yu Feng, Yuhan Jiang 외 arxiv

Given a map description through global traversal navigation instructions, an LLM can often infer the implicit spatial layout and answer user queries by providing shortest paths. However, such context-dependent querying b…

Comparative Evaluation of the Proximate and Cytogenotoxicity of Ash and Rice Chips Used as Mango Fruit Artificial Ripening Agents in Birnin Kebbi, Nigeria

2024-06-29 · CD Obadiah, TO Yahaya, AA Aliero, M Abdulkareem

The high demand for mango (Mangifera indica L.) fruits has led sellers to employ ripening agents. However, concerns are growing regarding the potential toxicities of induced ripening, emphasizing the need for scientific …

MangoLeafViT: Leveraging Lightweight Vision Transformer with Runtime Augmentation for Efficient Mango Leaf Disease Classification

2025-05-29 · Rafi Hassan Chowdhury, Sabbir Ahmed

Ensuring food safety is critical due to its profound impact on public health, economic stability, and global supply chains. Cultivation of Mango, a major agricultural product in several South Asian countries, faces high …