Demystifying Instruction Mixing for Fine-tuning Large Language Models
This work addresses the problem of optimizing instruction mixing for LLM fine-tuning, which is incremental as it builds on existing tuning methods.
The study investigated how mixing different types of instruction datasets (NLP tasks, coding, and general chat) affects fine-tuning performance of large language models, finding that certain instruction types benefit specific applications but can harm others.
Instruction tuning significantly enhances the performance of large language models (LLMs) across various tasks. However, the procedure to optimizing the mixing of instruction datasets for LLM fine-tuning is still poorly understood. This study categorizes instructions into three primary types: NLP downstream tasks, coding, and general chat. We explore the effects of instruction tuning on different combinations of datasets on LLM performance, and find that certain instruction types are more advantageous for specific applications but can negatively impact other areas. This work provides insights into instruction mixtures, laying the foundations for future research.