Do LLMs Encode Frame Semantics? Evidence from Frame Identification
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909989677826048 |
|---|---|
| author | Chundru, Jayanth Krishna Poddar, Rudrashis Cao, Jie Jiang, Tianyu |
| author_facet | Chundru, Jayanth Krishna Poddar, Rudrashis Cao, Jie Jiang, Tianyu |
| contents | We investigate whether large language models encode latent knowledge of frame semantics, focusing on frame identification, a core challenge in frame semantic parsing that involves selecting the appropriate semantic frame for a target word in context. Using the FrameNet lexical resource, we evaluate models under prompt-based inference and observe that they can perform frame identification effectively even without explicit supervision. To assess the impact of task-specific training, we fine-tune the model on FrameNet data, which substantially improves in-domain accuracy while generalizing well to out-of-domain benchmarks. Further analysis shows that the models can generate semantically coherent frame definitions, highlighting the model's internalized understanding of frame semantics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_19540 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Do LLMs Encode Frame Semantics? Evidence from Frame Identification Chundru, Jayanth Krishna Poddar, Rudrashis Cao, Jie Jiang, Tianyu Computation and Language We investigate whether large language models encode latent knowledge of frame semantics, focusing on frame identification, a core challenge in frame semantic parsing that involves selecting the appropriate semantic frame for a target word in context. Using the FrameNet lexical resource, we evaluate models under prompt-based inference and observe that they can perform frame identification effectively even without explicit supervision. To assess the impact of task-specific training, we fine-tune the model on FrameNet data, which substantially improves in-domain accuracy while generalizing well to out-of-domain benchmarks. Further analysis shows that the models can generate semantically coherent frame definitions, highlighting the model's internalized understanding of frame semantics. |
| title | Do LLMs Encode Frame Semantics? Evidence from Frame Identification |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.19540 |