Distilling Instruction-following Abilities of Large Language Models with Task-aware Curriculum Planning

The process of instruction tuning aligns pre-trained large language models (LLMs) with open-domain instructions and human-preferred responses. While several studies have explored autonomous approaches to distilling and annotating instructions from more powerful proprietary LLMs, such as ChatGPT, they often neglect the impact of task distributions and the varying difficulty of instructions of the training sets. This oversight can lead to imbalanced knowledge capabilities and poor generalization powers of small student LLMs. To address this challenge, we introduce Task-Aware Curriculum Planning for Instruction Refinement (TAPIR), a multi-round distillation framework with balanced <PRE_TAG>task distributions</POST_TAG> and dynamic difficulty adjustment. This approach utilizes an oracle LLM to select instructions that are difficult for a student LLM to follow and distill instructions with balanced <PRE_TAG>task distributions</POST_TAG>. By incorporating curriculum planning, our approach systematically escalates the difficulty levels, progressively enhancing the student LLM's capabilities. We rigorously evaluate TAPIR using two widely recognized benchmarks, including AlpacaEval 2.0 and MT-Bench. The empirical results demonstrate that the student LLMs, trained with our method and less training data, outperform larger instruction-tuned models and strong distillation baselines. The improvement is particularly notable in complex tasks, such as logical reasoning and code generation.