STRUCTUREDAGENT: Planning with AND/OR Trees for Long-Horizon Web Tasks
#STRUCTUREDAGENT #AND/OR trees #long-horizon tasks #web automation #task planning
📌 Key Takeaways
- STRUCTUREDAGENT is a new AI system for complex web tasks using AND/OR trees.
- It focuses on long-horizon planning to handle multi-step online activities.
- The approach improves efficiency and reliability in automated web interactions.
- Research addresses challenges in task decomposition and execution sequencing.
📖 Full Retelling
arXiv:2603.05294v1 Announce Type: new
Abstract: Recent advances in large language models (LLMs) have enabled agentic systems for sequential decision-making. Such agents must perceive their environment, reason across multiple time steps, and take actions that optimize long-term objectives. However, existing web agents struggle on complex, long-horizon tasks due to limited in-context memory for tracking history, weak planning abilities, and greedy behaviors that lead to premature termination. To
🏷️ Themes
AI Planning, Web Automation
Entity Intersection Graph
No entity connections available yet for this article.
Original Source
--> Computer Science > Artificial Intelligence arXiv:2603.05294 [Submitted on 5 Mar 2026] Title: STRUCTUREDAGENT: Planning with AND/OR Trees for Long-Horizon Web Tasks Authors: ELita Lobo , Xu Chen , Jingjing Meng , Nan Xi , Yang Jiao , Chirag Agarwal , Yair Zick , Yan Gao View a PDF of the paper titled STRUCTUREDAGENT: Planning with AND/OR Trees for Long-Horizon Web Tasks, by ELita Lobo and 7 other authors View PDF HTML Abstract: Recent advances in large language models have enabled agentic systems for sequential decision-making. Such agents must perceive their environment, reason across multiple time steps, and take actions that optimize long-term objectives. However, existing web agents struggle on complex, long-horizon tasks due to limited in-context memory for tracking history, weak planning abilities, and greedy behaviors that lead to premature termination. To address these challenges, we propose STRUCTUREDAGENT, a hierarchical planning framework with two core components: (1) an online hierarchical planner that uses dynamic AND/OR trees for efficient search and (2) a structured memory module that tracks and maintains candidate solutions to improve constraint satisfaction in information-seeking tasks. The framework also produces interpretable hierarchical plans, enabling easier debugging and facilitating human intervention when needed. Our results on WebVoyager, WebArena, and custom shopping benchmarks show that STRUCTUREDAGENT improves performance on long-horizon web-browsing tasks compared to standard LLM-based agents. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2603.05294 [cs.AI] (or arXiv:2603.05294v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2603.05294 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Elita Lobo [ view email ] [v1] Thu, 5 Mar 2026 15:37:06 UTC (2,061 KB) Full-text links: Access Paper: View a PDF of the paper titled STRUCTUREDAGENT: Planning with AND/OR Trees for Lo...
Read full article at source