Star-Agents: Automatic Data Optimization with LLM Agents for Instruction Tuning

TL;DR

This paper introduces Star-Agents, a framework that automates data quality enhancement for instruction tuning of LLMs through multi-agent collaboration, evaluation, and dynamic refinement, leading to significant performance improvements.

Contribution

The paper presents a novel multi-agent framework that automates data generation, assessment, and refinement to improve instruction tuning datasets for LLMs.

Findings

Achieved an average 12% performance increase on instruction tuning tasks.

Realized a 40% improvement in Fermi benchmark scores.

Demonstrated effectiveness across models like Pythia and LLaMA.

Abstract

The efficacy of large language models (LLMs) on downstream tasks usually hinges on instruction tuning, which relies critically on the quality of training data. Unfortunately, collecting high-quality and diverse data is both expensive and time-consuming. To mitigate this issue, we propose a novel Star-Agents framework, which automates the enhancement of data quality across datasets through multi-agent collaboration and assessment. The framework adopts a three-pronged strategy. It initially generates diverse instruction data with multiple LLM agents through a bespoke sampling method. Subsequently, the generated data undergo a rigorous evaluation using a dual-model method that assesses both difficulty and quality. Finaly, the above process evolves in a dynamic refinement phase, where more effective LLMs are prioritized, enhancing the overall data quality. Our empirical studies, including instruction tuning experiments with models such as Pythia and LLaMA, demonstrate the effectiveness of the proposed framework. Optimized datasets have achieved substantial improvements, with an average increase of 12% and notable gains in specific metrics, such as a 40% improvement in Fermi, as evidenced by benchmarks like MT-bench, Vicuna bench, and WizardLM testset.