A framework that uses large language models to translate abstract dictionary definitions into concrete, illustrative visual scene descriptions.
Question
A dictionary definition is a compression. Vocabulary acquisition often fails at the last mile: the learner can recite the gloss and still cannot see the thing. Can generative models reverse that compression into a scene that carries the sense?
Constraint
The picture is accountable to the definition. Illustration is not decoration; it is a test of whether the model has the meaning or only the words around it.
Method
A computational framework that prompts large language models to expand abstract dictionary definitions into concrete visual scene descriptions, then studies whether those contextual visual narratives improve vocabulary acquisition and semantic comprehension.
Artifact
A generative lexical-visual pipeline for turning definitions into illustrative scenes.
Investigates efficacy of generative visual narrative as a semantic aid, not as clip-art.