Enhancing and Assessing Instruction-Following with Fine-Grained Instruction Variants

TL;DR

This paper introduces DeMoRecon, a data augmentation method that creates fine-grained instruction variants to improve LLMs' ability to follow instructions accurately, and presents the FGIV dataset for training and evaluation.

Contribution

The paper proposes DeMoRecon for generating instruction variants and introduces the FGIV dataset, enhancing instruction-following assessment and training of LLMs.

Findings

LLMs fine-tuned with FGIV show significant performance improvements.

DeMoRecon effectively preserves instruction context while increasing variability.

FGIV enhances both training and evaluation of instruction-following capabilities.

Abstract

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.