Xinbo Wu


2024

pdf bib
A Meta-Learning Perspective on Transformers for Causal Language Modeling
Xinbo Wu | Lav Varshney
Findings of the Association for Computational Linguistics ACL 2024

The Transformer architecture has become prominent in developing large causal language models. However, mechanisms to explain its capabilities are not well understood. Focused on the training process, here we establish a meta-learning view of the Transformer architecture when trained for the causal language modeling task, by explicating an inner optimization process that may happen within the Transformer. Further, from within the inner optimization, we discover and theoretically analyze a special characteristic of the norms of learned token representations within Transformer-based causal language models. Our analysis is supported by experiments conducted on pre-trained large language models and real-world data.
Search
Co-authors
Venues