Let Me Speak Freely? A Study On The Impact Of Format Restrictions On Performance Of Large Language Models
Tam Zhi Rui, Wu Cheng-kuang, Tsai Yi-lin, Lin Chieh-yen, Lee Hung-yi, Chen Yun-nung. Arxiv 2024
[Paper]
Structured generation, the process of producing content in standardized formats like JSON and XML, is widely utilized in real-world applications to extract key output information from large language models (LLMs). This study investigates whether such constraints on generation space impact LLMs’ abilities, including reasoning and domain knowledge comprehension. Specifically, we evaluate LLMs’ performance when restricted to adhere to structured formats versus generating free-form responses across various common tasks. Surprisingly, we observe a significant decline in LLMs’ reasoning abilities under format restrictions. Furthermore, we find that stricter format constraints generally lead to greater performance degradation in reasoning tasks.
Similar Work