Including copying tasks in training enables transformers to learn letter-string analogies, improving generalization to new alphabets with a 3-layer model outperforming some frontier models.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Transformer See, Transformer Do: Copying as an Intermediate Step in Learning Analogical Reasoning
Including copying tasks in training enables transformers to learn letter-string analogies, improving generalization to new alphabets with a 3-layer model outperforming some frontier models.