Diverse and Fine-Grained Instruction-Following Ability Exploration with Synthetic Data
This addresses the problem of limited evaluation methods for LLMs' instruction-following capabilities, offering a tool for researchers and developers, though it is incremental as it builds on existing evaluation frameworks.
The paper tackles the challenge of evaluating large language models' instruction-following abilities by introducing DINGO, a dataset with 130 fine-grained categories and diverse instructions, which provides more comprehensive and challenging evaluations.
Instruction-following is particularly crucial for large language models (LLMs) to support diverse user requests. While existing work has made progress in aligning LLMs with human preferences, evaluating their capabilities on instruction following remains a challenge due to complexity and diversity of real-world user instructions. While existing evaluation methods focus on general skills, they suffer from two main shortcomings, i.e., lack of fine-grained task-level evaluation and reliance on singular instruction expression. To address these problems, this paper introduces DINGO, a fine-grained and diverse instruction-following evaluation dataset that has two main advantages: (1) DINGO is based on a manual annotated, fine-grained and multi-level category tree with 130 nodes derived from real-world user requests; (2) DINGO includes diverse instructions, generated by both GPT-4 and human experts. Through extensive experiments, we demonstrate that DINGO can not only provide more challenging and comprehensive evaluation for LLMs, but also provide task-level fine-grained directions to further improve LLMs.