From Complex to Simple: Enhancing Multi-Constraint Complex Instruction Following Ability of Large Language Models

TL;DR

This paper investigates how training data with multiple constraints improves large language models' ability to follow complex instructions, proposing methods to enhance their understanding and generalization across various settings.

Contribution

It identifies effective training data strategies and introduces methods to improve LLMs' complex instruction-following and generalization capabilities.

Findings

Training with multi-constraint instructions enhances understanding of complex tasks.

Proposed methods improve performance and training efficiency.

Models generalize well across out-of-domain, in-domain, and adversarial scenarios.

Abstract

It is imperative for Large language models (LLMs) to follow instructions with elaborate requirements (i.e. Complex Instructions Following). Yet, it remains under-explored how to enhance the ability of LLMs to follow complex instructions with multiple constraints. To bridge the gap, we initially study what training data is effective in enhancing complex constraints following abilities. We found that training LLMs with instructions containing multiple constraints enhances their understanding of complex instructions, especially those with lower complexity levels. The improvement can even generalize to compositions of out-of-domain constraints. Additionally, we further propose methods addressing how to obtain and utilize the effective training data. Finally, we conduct extensive experiments to prove the effectiveness of our methods in terms of overall performance and training efficiency. We also demonstrate that our methods improve models' ability to follow instructions generally and generalize effectively across out-of-domain, in-domain, and adversarial settings, while maintaining general capabilities.