Zero Shot Phonics: Evaluating Constraint-Adherent Phonics Story Generation in Large Language Models
Year: 2026
Type: conference-paper
Abstract: Phonics stories are essential for early literacy, requiring controlled repetition of graphemephoneme (GP) patterns while remaining simple, readable, and developmentally appropriate.Generating such texts poses a challenge for large language models (LLMs), which must handle multiple linguistic and pedagogical constraints.To investigate how these constraints can be balanced, we systematically vary prompt design across 16 configurations and evaluate six LLMs in a zero-shot setting, producing 8,688 outputs (39,096 stories).Outputs are assessed using a multi-dimensional framework covering phonological alignment, developmental lexical appropriateness, readability, and narrative quality.Results show that while LLMs generate highly readable and age-appropriate text, they exhibit variability in phoneme control and narrative coherence.Prompt design and model choice significantly affect performance, revealing trade-offs across constraints.These findings highlight the challenges of controllable educational text generation and the importance of prompt design in balancing instructional objectives.We release our prompts, generated stories, and evaluation framework to support future work in phonics-based story generation.
Authors: Maria Monica Manlises, Ethel Ong
Institution: De La Salle University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: hybrid
OpenAlex ID: W7166830600