From Complex to Simple: Enhancing Multi-Constraint Complex Instruction Following Ability of Large Language Models
This work addresses the challenge of complex instruction following for large language models, which is incremental as it builds on existing methods to improve specific aspects of training data.
The paper tackles the problem of enhancing large language models' ability to follow complex instructions with multiple constraints by identifying effective training data and proposing methods to obtain and use it, resulting in improved performance and generalization across various settings while maintaining general capabilities.
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.