Abstractors and relational cross-attention: An inductive bias for explicit relational reasoning in Transformers

arXiv:2304.00195v4 Announce Type: replace-cross
Abstract: An extension of Transformers is proposed that enables explicit relational reasoning through a novel module called the Abstractor. At the core of the Abstractor is a variant of attention called relational cross-attention. The approach is motivated by an architectural inductive bias for relational learning that disentangles relational information from object-level features. This enables explicit relational reasoning, supporting abstraction and generalization from limited data. The Abstractor is first evaluated on simple discriminative relational tasks and compared …

abstract arxiv attention bias core cs.lg extension inductive novel reasoning relational stat.ml through transformers type