Enhancing and Assessing Instruction-Following with Fine-Grained Instruction Variants
This addresses the challenge of precise instruction alignment for LLMs in real-world applications, representing an incremental improvement over existing methods focused on sample diversity.
The paper tackles the problem of accurately assessing and improving Large Language Models' ability to follow similar instruction variants by introducing DeMoRecon, a data augmentation technique that decomposes, modifies, and reconstructs instructions to create fine-grained variants, resulting in a dataset of 1,773 seed instructions that significantly boosts performance on instruction-following benchmarks.
The effective alignment of Large Language Models (LLMs) with precise instructions is essential for their application in diverse real-world scenarios. Current methods focus on enhancing the diversity and complexity of training and evaluation samples, yet they fall short in accurately assessing LLMs' ability to follow similar instruction variants. We introduce an effective data augmentation technique DeMoRecon that decomposes complex instructions into simpler sub-components, modifies these, and reconstructs them into new variants, thereby preserves the original instruction's context and complexity while introducing variability, which is critical for training and evaluating LLMs' instruction-following precision. Based on DeMoRecon, we developed the FGIV dataset which contains fine-grained instruction variants of 1,773 seed instructions to both fine-tune and evaluate LLMs. Our findings show that LLMs fine-tuned with FGIV will gain significant performance boost on both ours and commonly used instructions-following benchmarks.