Search

Your search keyword '"Loppi, N"' showing total 4 results
Start Over Author "Loppi, N" Publication Year Range Last 3 years
4 results on '"Loppi, N"'

1. MAMMOTH Massively Multilingual Modular Open Translation @ Helsinki

Author

Mickus, T, Gronroos, S, Attieh, J, Boggia, M, De Gibert, O, Ji, S, Loppi, N, Raganato, A, Vazquez, R, Tiedemann, J, Mickus T., Gronroos S. -A., Attieh J., Boggia M., De Gibert O., Ji S., Loppi N. A., Raganato A., Vazquez R., Tiedemann J., Mickus, T, Gronroos, S, Attieh, J, Boggia, M, De Gibert, O, Ji, S, Loppi, N, Raganato, A, Vazquez, R, Tiedemann, J, Mickus T., Gronroos S. -A., Attieh J., Boggia M., De Gibert O., Ji S., Loppi N. A., Raganato A., Vazquez R., and Tiedemann J.

Abstract

NLP in the age of monolithic large language models is approaching its limits in terms of size and information that can be handled. The trend goes to modularization, a necessary step into the direction of designing smaller sub-networks and components with specialized functionality. In this paper, we present the MAMMOTH toolkit: a framework designed for training massively multilingual modular machine translation systems at scale, initially derived from OpenNMT-py and then adapted to ensure efficient training across computation clusters. We showcase its efficiency across clusters of A100 and V100 NVIDIA GPUs, and discuss our design philosophy and plans for future information. The toolkit is publicly available online.

Published
2024

3. Latest Development in the FoTran Project - Scaling Up Language Coverage in Neural Machine Translation Using Distributed Training with Language-Specific Components

Author

Macken, L, Rufener, A, an den Bogaert, J, Daems, J, Tezcan, A, Vanroy, B, Fonteyne, M, Barrault, L, Costa-Jussa, MR, Kemp, E, Pilos, E, Declercq, C, Koponen, M, Forcada, ML, Scarton, C, Moniz, H, Vazquez, R, Boggia, M, Raganato, A, Loppi, N, Gronroos, S, Tiedemann, J, Vazquez R., Boggia M., Raganato A., Loppi N. A., Gronroos S. -A., Tiedemann J., Macken, L, Rufener, A, an den Bogaert, J, Daems, J, Tezcan, A, Vanroy, B, Fonteyne, M, Barrault, L, Costa-Jussa, MR, Kemp, E, Pilos, E, Declercq, C, Koponen, M, Forcada, ML, Scarton, C, Moniz, H, Vazquez, R, Boggia, M, Raganato, A, Loppi, N, Gronroos, S, Tiedemann, J, Vazquez R., Boggia M., Raganato A., Loppi N. A., Gronroos S. -A., and Tiedemann J.

Abstract

We give an update on the Found in Translation (FoTran) project, focusing on the study of emerging language-agnostic representations from neural machine translation (NMT). We describe our attention-bridge model, a modular NMT model which connects language-specific components through a shared network layer. Our latest implementation supports distributed training over many nodes and GPUs in order to substantially scale up the number of languages that can be included in a modern neural translation architecture.

Published
2022

4. Latest Development in the FoTran Project - Scaling Up Language Coverage in Neural Machine Translation Using Distributed Training with Language-Specific Components

Author

Vazquez R., Boggia M., Raganato A., Loppi N. A., Gronroos S. -A., Tiedemann J., Macken, L, Rufener, A, an den Bogaert, J, Daems, J, Tezcan, A, Vanroy, B, Fonteyne, M, Barrault, L, Costa-Jussa, MR, Kemp, E, Pilos, E, Declercq, C, Koponen, M, Forcada, ML, Scarton, C, Moniz, H, Vazquez, R, Boggia, M, Raganato, A, Loppi, N, Gronroos, S, and Tiedemann, J

Subjects
Program processor, Shared network, Neural machine translation, Scale-up, Computational linguistic, Implementation support, Specific component, Machine translation model, Machine component, Modular, Bridge model, Latest development, Scaling-up, Network layer, Computer aided language translation
Abstract

We give an update on the Found in Translation (FoTran) project, focusing on the study of emerging language-agnostic representations from neural machine translation (NMT). We describe our attention-bridge model, a modular NMT model which connects language-specific components through a shared network layer. Our latest implementation supports distributed training over many nodes and GPUs in order to substantially scale up the number of languages that can be included in a modern neural translation architecture.

Published
2022

Catalog

Books, media, physical & digital resources
See catalog results