flâneur

The Topological Trouble With Transformers

arxiv.org · 9,450 words · saved by 1 readers

Transformers encode structure in sequences via an expanding contextual history. However, their purely feedforward architecture fundamentally limits dynamic state tracking. State tracking—the iterative updating of latent variables reflecting an evolving environment—involves inherently sequential dependencies that feedforward networks struggle to maintain. Consequently, feedforward models push evolving state representations deeper into their layer stack with each new input step, rendering information inaccessible in shallow layers and ultimately exhausting the model’s depth. While this depth limit can be bypassed by dynamic depth models and by explicit or latent thinking that externalizes state representations, these solutions are computationally and memory inefficient. In this article, we argue that temporally extended cognition requires refocusing from explicit thought traces to implicit activation dynamics via recurrent architectures. We introduce a taxonomy of recurrent and continuou

The Topological Trouble With Transformers Michael C. Mozer Google DeepMind mcmozer@google.com &Shoaib Ahmed Siddiqui Google DeepMind shoaibasidd@google.com &Rosanne Liu Google DeepMind rosanneliu@google.com Abstract Transformers encode structure in sequences via an expanding contextual history. However, their purely feedforward architecture fundamentally limits dynamic state tracking. State tracking—the iterative updating of latent variables reflecting an evolving environment—involves inherently sequential dependencies that feedforward networks struggle to maintain. Consequently, feedforward m

saved by

related reading