Can I read 2025 Acl-Long 1575 on EtoBox?

2025 Acl-Long 1575 by Minh Thương Nguyễn is a document available to read on EtoBox.

What is 2025 Acl-Long 1575 about?

The document presents H I AGENT, a framework designed to enhance the performance of Large Language Model (LLM)-based agents in long-horizon tasks by utilizing hierarchical working memory management through subgoals. This approach improves the success rate of agents by twofold and reduces the average number of steps required to complete tasks by 3.8, demonstrating significant efficiency gains compared to traditional methods. Experimental results indicate that H I AGENT effectively manages memory by summarizi

Author

Minh Thương Nguyễn

Language

EN