HyGenar: An LLM-Driven Hybrid Genetic Algorithm for Few-Shot Grammar Generation
This addresses the challenge of generating grammars for NLP and text/code generation tasks, but it is incremental as it builds on existing LLM and genetic algorithm methods.
The paper tackled the problem of few-shot grammar generation from examples using large language models (LLMs), finding that existing LLMs perform sub-optimally, and proposed HyGenar, a hybrid genetic algorithm that achieved substantial improvements in syntactic and semantic correctness.
Grammar plays a critical role in natural language processing and text/code generation by enabling the definition of syntax, the creation of parsers, and guiding structured outputs. Although large language models (LLMs) demonstrate impressive capabilities across domains, their ability to infer and generate grammars has not yet been thoroughly explored. In this paper, we aim to study and improve the ability of LLMs for few-shot grammar generation, where grammars are inferred from sets of a small number of positive and negative examples and generated in Backus-Naur Form. To explore this, we introduced a novel dataset comprising 540 structured grammar generation challenges, devised 6 metrics, and evaluated 8 various LLMs against it. Our findings reveal that existing LLMs perform sub-optimally in grammar generation. To address this, we propose an LLM-driven hybrid genetic algorithm, namely HyGenar, to optimize grammar generation. HyGenar achieves substantial improvements in both the syntactic and semantic correctness of generated grammars across LLMs.