Sökning: id:"swepub:oai:DiVA.org:hj-54862" >
How far can we get ...
How far can we get with one GPU in 100 hours? CoAStaL at MultiIndicMT Shared Task
-
Aralikatte, Rahul (författare)
-
Murrieta Bello, Héctor Ricardo (författare)
-
de Lhoneux, Miryam (författare)
-
visa fler...
-
Hershcovich, Daniel (författare)
-
- Bollmann, Marcel (författare)
- Jönköping University,Jönköping AI Lab (JAIL)
-
Søgaard, Anders (författare)
-
visa färre...
-
(creator_code:org_t)
- Stroudsburg, PA, USA : Association for Computational Linguistics, 2021
- 2021
- Engelska.
-
Ingår i: Proceedings of the 8th Workshop on Asian Translation (WAT2021). - Stroudsburg, PA, USA : Association for Computational Linguistics. ; , s. 205-211
- Relaterad länk:
-
https://aclanthology...
-
visa fler...
-
https://aclanthology...
-
https://urn.kb.se/re...
-
https://doi.org/10.1...
-
visa färre...
Abstract
Ämnesord
Stäng
- This work shows that competitive translation results can be obtained in a constrained setting by incorporating the latest advances in memory and compute optimization. We train and evaluate large multilingual translation models using a single GPU for a maximum of 100 hours and get within 4-5 BLEU points of the top submission on the leaderboard. We also benchmark standard baselines on the PMI corpus and re-discover well-known shortcomings of translation systems and metrics.
Ämnesord
- NATURVETENSKAP -- Data- och informationsvetenskap -- Språkteknologi (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Language Technology (hsv//eng)
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)