GOAT: A Training Framework for Goal-Oriented Agent with Tools
This addresses the challenge of building robust open-source LLM agents for complex reasoning and tool use, offering a practical solution for developers and researchers, though it is incremental as it builds on existing agent frameworks.
The paper tackles the problem of limited goal-oriented reasoning in LLM agents by proposing GOAT, a training framework that automatically constructs synthetic datasets from API documents, enabling fine-tuning without human annotation. The result is state-of-the-art performance on existing benchmarks and a new benchmark, GOATBench, with agents excelling in complex tool use.
Large language models (LLMs) have recently been extended beyond traditional text generation to serve as interactive agents capable of using external tools based on user intent. However, current LLM agents still show limited ability to handle goal-oriented queries, which require decomposing a high-level objective into multiple interdependent API calls with correct planning and execution. Current approaches mainly rely on zero-shot evaluation due to the absence of training data. While proprietary closed-source models such as GPT-4 demonstrate strong reasoning abilities, smaller open-source models struggle to perform complex tool use effectively. Thus, we propose a novel training framework GOAT, which enables fine-tuning of LLM agents in a human annotation-free setting. GOAT automatically constructs synthetic datasets of goal-oriented API execution tasks directly from given API documents, equipping models with the ability to reason over interdependent calls and generate coherent responses. Through extensive experiments, we show that GOAT-trained agents achieve state-of-the-art performance across multiple existing goal-oriented benchmarks. In addition, we introduce GOATBench, a new goal-oriented API execution benchmark, and demonstrate that agents trained with GOAT also excel in this setting. These results highlight GOAT as a practical path toward building robust open-source LLM agents capable of complex reasoning and tool use.