SelfGoal: Your Language Agents Already Know How to Achieve High-level Goals

TL;DR

SelfGoal is an innovative method that enables language agents to autonomously decompose high-level goals into subgoals, improving their ability to achieve complex objectives in environments with limited feedback.

Contribution

The paper introduces SelfGoal, a novel automatic approach for language agents to adaptively break down high-level goals into subgoals during interaction, enhancing goal achievement without extensive prior knowledge.

Findings

SelfGoal improves agent performance across diverse tasks.

It effectively handles environments with delayed or sparse feedback.

The approach outperforms baseline methods in goal achievement.

Abstract

Language agents powered by large language models (LLMs) are increasingly valuable as decision-making tools in domains such as gaming and programming. However, these agents often face challenges in achieving high-level goals without detailed instructions and in adapting to environments where feedback is delayed. In this paper, we present SelfGoal, a novel automatic approach designed to enhance agents' capabilities to achieve high-level goals with limited human prior and environmental feedback. The core concept of SelfGoal involves adaptively breaking down a high-level goal into a tree structure of more practical subgoals during the interaction with environments while identifying the most useful subgoals and progressively updating this structure. Experimental results demonstrate that SelfGoal significantly enhances the performance of language agents across various tasks, including competitive, cooperative, and deferred feedback environments. Project page: https://selfgoal-agent.github.io.