Different procedural pretraining tasks create complementary, transferable structures in a transformer's attention and MLP weights, and structures from different tasks can be combined into one initialization.
Pretraining with Artificial Language: Studying Transferable Knowledge in Language Models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
We investigate what kind of structural knowledge learned in neural network encoders is transferable to processing natural language. We design artificial languages with structural properties that mimic natural language, pretrain encoders on the data, and see how much performance the encoder exhibits on downstream tasks in natural language. Our experimental results show that pretraining with an artificial language with a nesting dependency structure provides some knowledge transferable to natural language. A follow-up probing analysis indicates that its success in the transfer is related to the amount of encoded contextual information and what is transferred is the knowledge of position-aware context dependence of language. Our results provide insights into how neural network encoders process human languages and the source of cross-lingual transferability of recent multilingual language models.
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Transformers Pretrained on Procedural Data Contain Modular Structures for Algorithmic Reasoning
Different procedural pretraining tasks create complementary, transferable structures in a transformer's attention and MLP weights, and structures from different tasks can be combined into one initialization.