no code implementations • WMT (EMNLP) 2021 • Lukas Edman, Ahmet Üstün, Antonio Toral, Gertjan van Noord
This paper describes the methods behind the systems submitted by the University of Groningen for the WMT 2021 Unsupervised Machine Translation task for German–Lower Sorbian (DE–DSB): a high-resource language to a low-resource one.
no code implementations • CL (ACL) 2022 • Ahmet Üstün, Arianna Bisazza, Gosse Bouma, Gertjan van Noord
To address this, we propose a novel language adaptation approach by introducing contextual language adapters to a multilingual parser.
no code implementations • 22 Feb 2024 • Arash Ahmadian, Chris Cremer, Matthias Gallé, Marzieh Fadaee, Julia Kreutzer, Olivier Pietquin, Ahmet Üstün, Sara Hooker
AI alignment in the shape of Reinforcement Learning from Human Feedback (RLHF) is increasingly treated as a crucial ingredient for high performance large language models.
no code implementations • 12 Feb 2024 • Ahmet Üstün, Viraat Aryabumi, Zheng-Xin Yong, Wei-Yin Ko, Daniel D'souza, Gbemileke Onilude, Neel Bhandari, Shivalika Singh, Hui-Lee Ooi, Amr Kayid, Freddie Vargus, Phil Blunsom, Shayne Longpre, Niklas Muennighoff, Marzieh Fadaee, Julia Kreutzer, Sara Hooker
Recent breakthroughs in large language models (LLMs) have centered around a handful of data-rich languages.
no code implementations • 9 Feb 2024 • Shivalika Singh, Freddie Vargus, Daniel Dsouza, Börje F. Karlsson, Abinaya Mahendiran, Wei-Yin Ko, Herumb Shandilya, Jay Patel, Deividas Mataciunas, Laura OMahony, Mike Zhang, Ramith Hettiarachchi, Joseph Wilson, Marina Machado, Luisa Souza Moura, Dominik Krzemiński, Hakimeh Fadaei, Irem Ergün, Ifeoma Okoh, Aisha Alaagib, Oshan Mudannayake, Zaid Alyafeai, Vu Minh Chien, Sebastian Ruder, Surya Guthikonda, Emad A. Alghamdi, Sebastian Gehrmann, Niklas Muennighoff, Max Bartolo, Julia Kreutzer, Ahmet Üstün, Marzieh Fadaee, Sara Hooker
The Aya initiative also serves as a valuable case study in participatory research, involving collaborators from 119 countries.
1 code implementation • 11 Sep 2023 • Ted Zadouri, Ahmet Üstün, Arash Ahmadian, Beyza Ermiş, Acyr Locatelli, Sara Hooker
The Mixture of Experts (MoE) is a widely known neural architecture where an ensemble of specialized sub-models optimizes overall performance with a constant computational cost.
no code implementations • 8 Sep 2023 • Max Marion, Ahmet Üstün, Luiza Pozzobon, Alex Wang, Marzieh Fadaee, Sara Hooker
In this work, we take a wider view and explore scalable estimates of data quality that can be used to systematically measure the quality of pretraining data.
1 code implementation • 24 May 2022 • Ahmet Üstün, Arianna Bisazza, Gosse Bouma, Gertjan van Noord, Sebastian Ruder
Massively multilingual models are promising for transfer learning across tasks and languages.
1 code implementation • 23 May 2022 • Ahmet Üstün, Asa Cooper Stickland
We find that using PEFTs with a larger pre-trained model outperforms full fine-tuning with a smaller model, and for smaller training data sizes, PEFTs outperform full fine-tuning for the same pre-trained model.
no code implementations • EMNLP 2021 • Ahmet Üstün, Alexandre Bérard, Laurent Besacier, Matthias Gallé
We consider the problem of multilingual unsupervised machine translation, translating to and from languages that only have monolingual data by using auxiliary parallel language pairs.
1 code implementation • 24 Sep 2021 • Lukas Edman, Ahmet Üstün, Antonio Toral, Gertjan van Noord
Lastly, we experiment with the order in which offline and online back-translation are used to train an unsupervised system, finding that using online back-translation first works better for DE$\rightarrow$DSB by 2. 76 BLEU.
2 code implementations • 13 Jul 2021 • Arianna Bisazza, Ahmet Üstün, Stephan Sportel
Identifying factors that make certain languages harder to model than others is essential to reach language equality in future Natural Language Processing technologies.
2 code implementations • NAACL 2021 • Rob van der Goot, Ibrahim Sharaf, Aizhan Imankulova, Ahmet Üstün, Marija Stepanović, Alan Ramponi, Siti Oryza Khairunnisa, Mamoru Komachi, Barbara Plank
To tackle the challenge, we propose a joint learning approach, with English SLU training data and non-English auxiliary tasks from raw text, syntax and translation for transfer.
no code implementations • EACL (AdaptNLP) 2021 • Rob van der Goot, Ahmet Üstün, Barbara Plank
However, it remains unclear in which situations these dataset embeddings are most effective, because they are used in a large variety of settings, languages and tasks.
1 code implementation • SEMEVAL 2020 • Bertelt Braaksma, Richard Scholtens, Stan van Suijlekom, Remy Wang, Ahmet Üstün
In this paper, we present our approach for sentiment classification on Spanish-English code-mixed social media data in the SemEval-2020 Task 9.
2 code implementations • EACL 2021 • Rob van der Goot, Ahmet Üstün, Alan Ramponi, Ibrahim Sharaf, Barbara Plank
In this paper we present MaChAmp, a toolkit for easy fine-tuning of contextualized embeddings in multi-task settings.
1 code implementation • EMNLP 2020 • Ahmet Üstün, Arianna Bisazza, Gosse Bouma, Gertjan van Noord
The resulting parser, UDapter, outperforms strong monolingual and multilingual baselines on the majority of both high-resource and low-resource (zero-shot) languages, showing the success of the proposed adaptation approach.
no code implementations • 24 Apr 2017 • Murathan Kurfali, Ahmet Üstün, Burcu Can
Our results show that using different information sources such as neural word embeddings and letter successor variety as prior information improves morphological segmentation in a Bayesian model.
no code implementations • 9 Mar 2017 • Burcu Can, Ahmet Üstün, Murathan Kurfali
We learn inflectional and derivational morpheme tags in Turkish by using conditional random fields (CRF) and we employ the morpheme tags in part-of-speech (PoS) tagging by using hidden Markov models (HMMs) to mitigate sparsity.