17 citations · 45 across the 12 of their papers we have counts for
Showing 2021Show all
2 papers · 1 filter
cs.CL2021
How Suitable Are Subword Segmentation Strategies for Translating Non-Concatenative Morphology?
Chantal Amrhein, Rico Sennrich
Data-driven subword segmentation has become the default strategy for open-vocabulary machine translation and other NLP tasks, but may not be sufficiently generic for optimal learni…
cs.CL2021
On Biasing Transformer Attention Towards Monotonicity
Annette Rios, Chantal Amrhein, Noëmi Aepli +1
Many sequence-to-sequence tasks in natural language processing are roughly monotonic in the alignment between source and target sequence, and previous work has facilitated or enfor…