Phun-Bench: Evaluating LLMs on Phonological Understanding in Chinese
For researchers studying LLM capabilities, this benchmark reveals a gap in phonological understanding that is not captured by existing benchmarks.
Phun-Bench evaluates LLMs on Chinese phonological understanding across homophony, rhyme, and phonetic similarity, finding that LLMs excel at pronunciation recall but struggle with flexible, human-like phonological reasoning.
Language is a vehicle for thought, intricately tied to sounds, symbols, and meaning. However, most large language model (LLM) research focuses on meaning (semantics) and symbols (spelling) while largely overlooking sounds. Existing benchmarks on LLMs' phonological abilities are either solvable through rote memorization or intertwined with other abilities, making them inadequate to measure LLMs' genuine ability in phonological understanding. Here, we present Phun-Bench, a purpose-built Chinese benchmark with diverse tasks and settings across three dimensions (Homophony, Rhyme, and Phonetic Similarity), designed to systematically evaluate LLMs' phonological understanding. Our results show that while LLMs excel at recalling correct pronunciations, they generally struggle to leverage phonological knowledge in the flexible and intuitive way that human speakers do. Moreover, through detailed analyses, we propose a hypothesis regarding the underlying mechanism of LLMs' phonological understanding and "perception", highlighting an underexplored frontier for future research.