Publications

Conference Papers

On the Representational Capacity of Recurrent Neural Language Models

An investigation into the formal languages and types of distributions that can be expressed by language models based on RNNs under different assumptions.

Franz Nowak, Anej Svete, Li Du, and Ryan Cotterell. 2023. On the Representational Capacity of Recurrent Neural Language Models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP), Singapore.

A Fast Algorithm for Computing Prefix Probabilities

This paper proposes a speedup of a classic algorithm for computing prefix probabilities under a probabilistic context-free grammar (PCFG).

Franz Nowak and Ryan Cotterell. 2023. A Fast Algorithm for Computing Prefix Probabilities. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL), Toronto, Canada.