Between Circuits and Chomsky: Pre-pretraining on Formal Languages Imparts Linguistic Biases
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_ | 1866912395787501568 |
|---|---|
| author | Hu, Michael Y. Petty, Jackson Shi, Chuan Merrill, William Linzen, Tal |
| author_facet | Hu, Michael Y. Petty, Jackson Shi, Chuan Merrill, William Linzen, Tal |
| contents | Pretraining language models on formal language can improve their acquisition of natural language. Which features of the formal language impart an inductive bias that leads to effective transfer? Drawing on insights from linguistics and complexity theory, we hypothesize that effective transfer occurs when two conditions are met: the formal language should capture the dependency structures present in natural language, and it should remain within the computational limitations of the model architecture. We experiment with pre-pretraining (training on formal language before natural languages) on transformers and find that formal languages capturing hierarchical dependencies indeed enable language models to achieve lower loss on natural language and better linguistic generalization compared to other formal languages. We also find modest support for the hypothesis that the formal language should fall within the computational limitations of the architecture. Strikingly, pre-pretraining reduces loss more efficiently than training on a matched amount of natural language. For a 1B-parameter language model trained on roughly 1.6B tokens of natural language, pre-pretraining achieves the same loss and better linguistic generalization with a 33% smaller token budget. Finally, we also give mechanistic evidence of transfer from formal to natural language: attention heads acquired during pre-pretraining remain crucial for the model's performance on syntactic evaluations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_19249 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Between Circuits and Chomsky: Pre-pretraining on Formal Languages Imparts Linguistic Biases Hu, Michael Y. Petty, Jackson Shi, Chuan Merrill, William Linzen, Tal Computation and Language Artificial Intelligence Machine Learning Pretraining language models on formal language can improve their acquisition of natural language. Which features of the formal language impart an inductive bias that leads to effective transfer? Drawing on insights from linguistics and complexity theory, we hypothesize that effective transfer occurs when two conditions are met: the formal language should capture the dependency structures present in natural language, and it should remain within the computational limitations of the model architecture. We experiment with pre-pretraining (training on formal language before natural languages) on transformers and find that formal languages capturing hierarchical dependencies indeed enable language models to achieve lower loss on natural language and better linguistic generalization compared to other formal languages. We also find modest support for the hypothesis that the formal language should fall within the computational limitations of the architecture. Strikingly, pre-pretraining reduces loss more efficiently than training on a matched amount of natural language. For a 1B-parameter language model trained on roughly 1.6B tokens of natural language, pre-pretraining achieves the same loss and better linguistic generalization with a 33% smaller token budget. Finally, we also give mechanistic evidence of transfer from formal to natural language: attention heads acquired during pre-pretraining remain crucial for the model's performance on syntactic evaluations. |
| title | Between Circuits and Chomsky: Pre-pretraining on Formal Languages Imparts Linguistic Biases |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2502.19249 |