LLMs struggle to follow diverse instructions in knowledge-intensive writing tasks, highlighting the need for improvement.
Current large language models struggle to effectively follow user instructions for revising long-form answers, as shown by the KIWI dataset analysis.
Current LLMs struggle to effectively follow expert-written instructions for revising long-form answers in the scientific domain.
The author presents KIWI, a dataset of expert-written instructions to revise long-form answers to research questions, highlighting the challenges faced by current LLMs in following instructions accurately.