← Search

Mark Schöne

2 accepted papers

2025

Implicit Language Models are RNNs: Balancing Parallelization and Expressivity

ICML 2025spotlight

State-space models (SSMs) and transformers dominate the language modeling landscape. However, they are constrained to a lower computational complexity than classical recurrent neural networks (RNNs), limiting their expressivity. In contrast, RNNs lack parallelization during training, raising fundame…

2023

Efficient recurrent architectures through activity sparsity and sparse back-propagation through time

ICLR 2023top-25%

Recurrent neural networks (RNNs) are well suited for solving sequence tasks in resource-constrained systems due to their expressivity and low computational requirements. However, there is still a need to bridge the gap between what RNNs are capable of in terms of efficiency and performance and real…