Long-Horizon Plan Execution in Large Tool Spaces through Entropy-Guided Branching

TL;DR

This paper introduces a new benchmark for evaluating large language model agents in tool-rich environments and proposes an entropy-guided search method to improve multi-step task execution efficiency.

Contribution

The paper presents SLATE, a large-scale benchmark for tool-augmented agents, and proposes Entropy-Guided Branching, a novel search algorithm that enhances exploration and efficiency.

Findings

Current agents struggle with self-correction and search efficiency on SLATE.

EGB significantly improves task success rates.

EGB reduces computational costs during planning.

Abstract

Large Language Models (LLMs) have significantly advanced tool-augmented agents, enabling autonomous reasoning via API interactions. However, executing multi-step tasks within massive tool libraries remains challenging due to two critical bottlenecks: (1) the absence of rigorous, plan-level evaluation frameworks and (2) the computational demand of exploring vast decision spaces stemming from large toolsets and long-horizon planning. To bridge these gaps, we first introduce SLATE (Synthetic Large-scale API Toolkit for E-commerce), a large-scale context-aware benchmark designed for the automated assessment of tool-integrated agents. Unlike static metrics, SLATE accommodates diverse yet functionally valid execution trajectories, revealing that current agents struggle with self-correction and search efficiency. Motivated by these findings, we next propose Entropy-Guided Branching (EGB), an uncertainty-aware search algorithm that dynamically expands decision branches where predictive entropy is high. EGB optimizes the exploration-exploitation trade-off, significantly enhancing both task success rates and computational efficiency. Extensive experiments on SLATE demonstrate that our dual contribution provides a robust foundation for developing reliable and scalable LLM agents in tool-rich environments.