Ahmet Üstün

67 papers A* 7Misc 1Journal 41Unranked 18
YearRankTypeTitle / Venue / Authors
2026 J jnl
Tiny Aya: Bridging Scale and Multilingual Depth.
CoRR
Alejandro Salamanca, Diana Abagyan, Daniel D'souza, Ammar Khairi, David Mora, Saurabh Dash, Viraat Aryabumi, Sara Rajaee, Mehrnaz Mofakhami, Ananya Sahu, Thomas Euyang, Brittawnya Prince, Madeline Smith, Hangyu Lin, Acyr Locatelli, Sara Hooker, Tom Kocmi, Aidan N. Gomez, Ivan Zhang, Phil Blunsom, Nick Frosst, Joelle Pineau, Beyza Ermis, Ahmet Üstün, Julia Kreutzer, Marzieh Fadaee
2025 J jnl
A Post-trainer's Guide to Multilingual Training Data: Uncovering Cross-lingual Transfer Dynamics.
CoRR
Luísa Shimabucoro, Ahmet Üstün, Marzieh Fadaee, Sebastian Ruder
2025 J jnl
Aya Vision: Advancing the Frontier of Multilingual Multimodality.
CoRR
Saurabh Dash, Yiyang Nan, John Dang, Arash Ahmadian, Shivalika Singh, Madeline Smith, Bharat Venkitesh, Vlad Shmyhlo, Viraat Aryabumi, Walter Beller-Morales, Jeremy Pekmez, Jason Ozuzu, Pierre Richemond, Acyr Locatelli, Nick Frosst, Phil Blunsom, Aidan N. Gomez, Ivan Zhang, Marzieh Fadaee, Manoj Govindassamy, Sudip Roy, Matthias Gallé, Beyza Ermis, Ahmet Üstün, Sara Hooker
2025 J jnl
EAGER: Entropy-Aware GEneRation for Adaptive Inference-Time Scaling.
CoRR
Daniel Scalena, Leonidas Zotos, Elisabetta Fersini, Malvina Nissim, Ahmet Üstün
2025 J jnl
MURI: High-Quality Instruction Tuning Datasets for Low-Resource Languages via Reverse Instructions.
Trans. Assoc. Comput. Linguistics
Abdullatif Köksal, Marion Thaler, Ayyoob Imani, Ahmet Üstün, Anna Korhonen, Hinrich Schütze
2025 — conf
Nexus: Adaptive Upcycling to Efficiently Pretrain Mixture of Experts.
EMNLP (Findings)
Nikolas Gritsch, Qizhen Zhang, Acyr Locatelli, Sara Hooker, Ahmet Üstün
2025 J jnl
One Tokenizer To Rule Them All: Emergent Language Plasticity via Multilingual Tokenizers.
CoRR
Diana Abagyan, Alejandro Salamanca, Felipe Cruz-Salinas, Kris Cao, Hangyu Lin, Acyr Locatelli, Marzieh Fadaee, Ahmet Üstün, Sara Hooker
2025 J jnl
The Disparate Impacts of Speculative Decoding.
CoRR
Jameson Sandler, Ahmet Üstün, Marco Romanelli, Sara Hooker, Ferdinando Fioretto
2025 J jnl
The Leaderboard Illusion.
CoRR
Shivalika Singh, Yiyang Nan, Alex Wang, Daniel D'souza, Sayash Kapoor, Ahmet Üstün, Sanmi Koyejo, Yuntian Deng, Shayne Longpre, Noah Smith, Beyza Ermis, Marzieh Fadaee, Sara Hooker
2025 J jnl
The Multilingual Divide and Its Impact on Global AI Safety.
CoRR
Aidan Peppin, Julia Kreutzer, Alice Schoenauer Sebag, Kelly Marchisio, Beyza Ermis, John Dang, Samuel Cahyawijaya, Shivalika Singh, Seraphina Goldfarb-Tarrant, Viraat Aryabumi, Aakanksha, Wei-Yin Ko, Ahmet Üstün, Matthias Gallé, Marzieh Fadaee, Sara Hooker
2025 A* conf
To Code or Not To Code? Exploring Impact of Code in Pre-training.
ICLR
Viraat Aryabumi, Yixuan Su, Raymond Ma, Adrien Morisot, Ivan Zhang, Acyr Locatelli, Marzieh Fadaee, Ahmet Üstün, Sara Hooker
2025 J jnl
Treasure Hunt: Real-time Targeting of the Long Tail using Training-Time Markers.
CoRR
Daniel D'souza, Julia Kreutzer, Adrien Morisot, Ahmet Üstün, Sara Hooker
2025 — conf
When Personalization Meets Reality: A Multi-Faceted Analysis of Personalized Preference Learning.
EMNLP (Findings)
Yijiang River Dong, Tiancheng Hu, Yinhong Liu, Ahmet Üstün, Nigel Collier
2025 J jnl
When Personalization Meets Reality: A Multi-Faceted Analysis of Personalized Preference Learning.
CoRR
Yijiang River Dong, Tiancheng Hu, Yinhong Liu, Ahmet Üstün, Nigel Collier
2024 J jnl
Aya 23: Open Weight Releases to Further Multilingual Progress.
CoRR
Viraat Aryabumi, John Dang, Dwarak Talupuru, Saurabh Dash, David Cairuz, Hangyu Lin, Bharat Venkitesh, Madeline Smith, Jon Ander Campos, Yi Chern Tan, Kelly Marchisio, Max Bartolo, Sebastian Ruder, Acyr Locatelli, Julia Kreutzer, Nick Frosst, Aidan N. Gomez, Phil Blunsom, Marzieh Fadaee, Ahmet Üstün, Sara Hooker
2024 — conf
Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning.
ACL (1)
Shivalika Singh, Freddie Vargus, Daniel D'souza, Börje Karlsson, Abinaya Mahendiran, Wei-Yin Ko, Herumb Shandilya, Jay Patel, Deividas Mataciunas, Laura O'Mahony, Mike Zhang, Ramith Hettiarachchi, Joseph Wilson, Marina Machado, Luisa Souza Moura, Dominik Krzeminski, Hakimeh Fadaei, Irem Ergün, Ifeoma Okoh, Aisha Alaagib, Oshan Mudannayake, Zaid Alyafeai, Minh Vu Chien, Sebastian Ruder, Surya Guthikonda, Emad A. Alghamdi, Sebastian Gehrmann, Niklas Muennighoff, Max Bartolo, Julia Kreutzer, Ahmet Üstün, Marzieh Fadaee, Sara Hooker
2024 J jnl
Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning.
CoRR
Shivalika Singh, Freddie Vargus, Daniel D'souza, Börje F. Karlsson, Abinaya Mahendiran, Wei-Yin Ko, Herumb Shandilya, Jay Patel, Deividas Mataciunas, Laura O'Mahony, Mike Zhang, Ramith Hettiarachchi, Joseph Wilson, Marina Machado, Luisa Souza Moura, Dominik Krzeminski, Hakimeh Fadaei, Irem Ergün, Ifeoma Okoh, Aisha Alaagib, Oshan Mudannayake, Zaid Alyafeai, Minh Chien Vu, Sebastian Ruder, Surya Guthikonda, Emad A. Alghamdi, Sebastian Gehrmann, Niklas Muennighoff, Max Bartolo, Julia Kreutzer, Ahmet Üstün, Marzieh Fadaee, Sara Hooker
2024 J jnl
Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontier.
CoRR
John Dang, Shivalika Singh, Daniel D'souza, Arash Ahmadian, Alejandro Salamanca, Madeline Smith, Aidan Peppin, Sungjin Hong, Manoj Govindassamy, Terrence Zhao, Sandra Kublik, Meor Amer, Viraat Aryabumi, Jon Ander Campos, Yi Chern Tan, Tom Kocmi, Florian Strub, Nathan Grinsztajn, Yannis Flet-Berliac, Acyr Locatelli, Hangyu Lin, Dwarak Talupuru, Bharat Venkitesh, David Cairuz, Bowen Yang, Tim Chung, Wei-Yin Ko, Sylvie Shang Shi, Amir Shukayev, Sammie Bae, Aleksandra Piktus, Roman Castagné, Felipe Cruz-Salinas, Eddie Kim, Lucas Crawhall-Stein, Adrien Morisot, Sudip Roy, Phil Blunsom, Ivan Zhang, Aidan N. Gomez, Nick Frosst, Marzieh Fadaee, Beyza Ermis, Ahmet Üstün, Sara Hooker
2024 — conf
Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model.
ACL (1)
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
2024 J jnl
Aya Model: An Instruction Finetuned Open-Access Multilingual Language Model.
CoRR
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
2024 A* conf
BAM! Just Like That: Simple and Efficient Parameter Upcycling for Mixture of Experts.
NeurIPS
Qizhen (Irene) Zhang, Nikolas Gritsch, Dwaraknath Gnaneshwar, Simon Guo, David Cairuz, Bharat Venkitesh, Jakob N. Foerster, Phil Blunsom, Sebastian Ruder, Ahmet Üstün, Acyr Locatelli
2024 J jnl
BAM! Just Like That: Simple and Efficient Parameter Upcycling for Mixture of Experts.
CoRR
Qizhen Zhang, Nikolas Gritsch, Dwaraknath Gnaneshwar, Simon Guo, David Cairuz, Bharat Venkitesh, Jakob N. Foerster, Phil Blunsom, Sebastian Ruder, Ahmet Üstün, Acyr Locatelli
2024 J jnl
Back to Basics: Revisiting REINFORCE Style Optimization for Learning from Human Feedback in LLMs.
CoRR
Arash Ahmadian, Chris Cremer, Matthias Gallé, Marzieh Fadaee, Julia Kreutzer, Olivier Pietquin, Ahmet Üstün, Sara Hooker
2024 — conf
Back to Basics: Revisiting REINFORCE-Style Optimization for Learning from Human Feedback in LLMs.
ACL (1)
Arash Ahmadian, Chris Cremer, Matthias Gallé, Marzieh Fadaee, Julia Kreutzer, Olivier Pietquin, Ahmet Üstün, Sara Hooker
2024 — conf
How Does Quantization Affect Multilingual LLMs?
EMNLP (Findings)
Kelly Marchisio, Saurabh Dash, Hongyu Chen, Dennis Aumiller, Ahmet Üstün, Sara Hooker, Sebastian Ruder
2024 J jnl
How Does Quantization Affect Multilingual LLMs?
CoRR
Kelly Marchisio, Saurabh Dash, Hongyu Chen, Dennis Aumiller, Ahmet Üstün, Sara Hooker, Sebastian Ruder
2024 J jnl
If You Can't Use Them, Recycle Them: Optimizing Merging at Scale Mitigates Performance Tradeoffs.
CoRR
Muhammad Khalifa, Yi Chern Tan, Arash Ahmadian, Tom Hosking, Honglak Lee, Lu Wang, Ahmet Üstün, Tom Sherborne, Matthias Gallé
2024 J jnl
MURI: High-Quality Instruction Tuning Datasets for Low-Resource Languages via Reverse Instructions.
CoRR
Abdullatif Köksal, Marion Thaler, Ayyoob Imani, Ahmet Üstün, Anna Korhonen, Hinrich Schütze
2024 J jnl
Nexus: Specialization meets Adaptability for Efficiently Training Mixture of Experts.
CoRR
Nikolas Gritsch, Qizhen Zhang, Acyr Locatelli, Sara Hooker, Ahmet Üstün
2024 A* conf
Pushing Mixture of Experts to the Limit: Extremely Parameter Efficient MoE for Instruction Tuning.
ICLR
Ted Zadouri, Ahmet Üstün, Arash Ahmadian, Beyza Ermis, Acyr Locatelli, Sara Hooker
2024 A* conf
RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMs.
EMNLP
John Dang, Arash Ahmadian, Kelly Marchisio, Julia Kreutzer, Ahmet Üstün, Sara Hooker
2024 J jnl
RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMs.
CoRR
John Dang, Arash Ahmadian, Kelly Marchisio, Julia Kreutzer, Ahmet Üstün, Sara Hooker
2024 J jnl
To Code, or Not To Code? Exploring Impact of Code in Pre-training.
CoRR
Viraat Aryabumi, Yixuan Su, Raymond Ma, Adrien Morisot, Ivan Zhang, Acyr Locatelli, Marzieh Fadaee, Ahmet Üstün, Sara Hooker
2023 A* conf
Intriguing Properties of Quantization at Scale.
NeurIPS
Arash Ahmadian, Saurabh Dash, Hongyu Chen, Bharat Venkitesh, Stephen Zhen Gou, Phil Blunsom, Ahmet Üstün, Sara Hooker
2023 J jnl
Intriguing Properties of Quantization at Scale.
CoRR
Arash Ahmadian, Saurabh Dash, Hongyu Chen, Bharat Venkitesh, Stephen Gou, Phil Blunsom, Ahmet Üstün, Sara Hooker
2023 J jnl
Pushing Mixture of Experts to the Limit: Extremely Parameter Efficient MoE for Instruction Tuning.
CoRR
Ted Zadouri, Ahmet Üstün, Arash Ahmadian, Beyza Ermis, Acyr Locatelli, Sara Hooker
2023 J jnl
When Less is More: Investigating Data Pruning for Pretraining LLMs at Scale.
CoRR
Max Marion, Ahmet Üstün, Luiza Pozzobon, Alex Wang, Marzieh Fadaee, Sara Hooker
2022 A* conf
Hyper-X: A Unified Hypernetwork for Multi-Task Multilingual Transfer.
EMNLP
Ahmet Üstün, Arianna Bisazza, Gosse Bouma, Gertjan van Noord, Sebastian Ruder
2022 J jnl
Hyper-X: A Unified Hypernetwork for Multi-Task Multilingual Transfer.
CoRR
Ahmet Üstün, Arianna Bisazza, Gosse Bouma, Gertjan van Noord, Sebastian Ruder
2022 J jnl
UDapter: Typology-based Language Adapters for Multilingual Dependency Parsing and Sequence Labeling.
Comput. Linguistics
Ahmet Üstün, Arianna Bisazza, Gosse Bouma, Gertjan van Noord
2022 A* conf
When does Parameter-Efficient Transfer Learning Work for Machine Translation?
EMNLP
Ahmet Üstün, Asa Cooper Stickland
2022 J jnl
When does Parameter-Efficient Transfer Learning Work for Machine Translation?
CoRR
Ahmet Üstün, Asa Cooper Stickland
2021 — conf
Automatic Judgement Forecasting for Pending Applications of the European Court of Human Rights.
ASAIL/LegalAIIA@ICAIL
Masha Medvedeva, Ahmet Üstün, Xiao Xu, Michel Vols, Martijn Wieling
2021 — conf
From Masked Language Modeling to Translation: Non-English Auxiliary Tasks Improve Zero-shot Spoken Language Understanding.
NAACL-HLT
Rob van der Goot, Ibrahim Sharaf, Aizhan Imankulova, Ahmet Üstün, Marija Stepanovic, Alan Ramponi, Siti Oryza Khairunnisa, Mamoru Komachi, Barbara Plank
2021 J jnl
From Masked Language Modeling to Translation: Non-English Auxiliary Tasks Improve Zero-shot Spoken Language Understanding.
CoRR
Rob van der Goot, Ibrahim Sharaf, Aizhan Imankulova, Ahmet Üstün, Marija Stepanovic, Alan Ramponi, Siti Oryza Khairunnisa, Mamoru Komachi, Barbara Plank
2021 J jnl
Incorporating word embeddings in unsupervised morphological segmentation.
Nat. Lang. Eng.
Ahmet Üstün, Burcu Can
2021 — conf
Massive Choice, Ample Tasks (MaChAmp): A Toolkit for Multi-task Learning in NLP.
EACL (System Demonstrations)
Rob van der Goot, Ahmet Üstün, Alan Ramponi, Ibrahim Sharaf, Barbara Plank
2021 — conf
Multilingual Unsupervised Neural Machine Translation with Denoising Adapters.
EMNLP (1)
Ahmet Üstün, Alexandre Berard, Laurent Besacier, Matthias Gallé
2021 J jnl
Multilingual Unsupervised Neural Machine Translation with Denoising Adapters.
CoRR
Ahmet Üstün, Alexandre Bérard, Laurent Besacier, Matthias Gallé
2021 J jnl
On the Difficulty of Translating Free-Order Case-Marking Languages.
CoRR
Arianna Bisazza, Ahmet Üstün, Stephan Sportel
2021 J jnl
On the Difficulty of Translating Free-Order Case-Marking Languages.
Trans. Assoc. Comput. Linguistics
Arianna Bisazza, Ahmet Üstün, Stephan Sportel
2021 J jnl
On the Effectiveness of Dataset Embeddings in Mono-lingual, Multi-lingual and Zero-shot Conditions.
CoRR
Rob van der Goot, Ahmet Üstün, Barbara Plank
2021 — conf
Unsupervised Translation of German-Lower Sorbian: Exploring Training and Novel Transfer Methods on a Low-Resource Language.
WMT@EMNLP
Lukas Edman, Ahmet Üstün, Antonio Toral, Gertjan van Noord
2021 J jnl
Unsupervised Translation of German-Lower Sorbian: Exploring Training and Novel Transfer Methods on a Low-Resource Language.
CoRR
Lukas Edman, Ahmet Üstün, Antonio Toral, Gertjan van Noord
2020 J jnl
FiSSA at SemEval-2020 Task 9: Fine-tuned For Feelings.
CoRR
Bertelt Braaksma, Richard Scholtens, Stan van Suijlekom, Remy Wang, Ahmet Üstün
2020 — conf
FiSSA at SemEval-2020 Task 9: Fine-tuned for Feelings.
SemEval@COLING
Bertelt Braaksma, Richard Scholtens, Stan van Suijlekom, Remy Wang, Ahmet Üstün
2020 J jnl
Massive Choice, Ample Tasks (MaChAmp): A Toolkit for Multi-task Learning in NLP.
CoRR
Rob van der Goot, Ahmet Üstün, Alan Ramponi, Barbara Plank
2020 — conf
UDapter: Language Adaptation for Truly Universal Dependency Parsing.
EMNLP (1)
Ahmet Üstün, Arianna Bisazza, Gosse Bouma, Gertjan van Noord
2020 J jnl
UDapter: Language Adaptation for Truly Universal Dependency Parsing.
CoRR
Ahmet Üstün, Arianna Bisazza, Gosse Bouma, Gertjan van Noord
2019 Misc conf
Cross-Lingual Word Embeddings for Morphologically Rich Languages.
RANLP
Ahmet Üstün, Gosse Bouma, Gertjan van Noord
2019 — conf
There and Back Again: Cross-Lingual Transfer Learning for Event Detection.
CLiC-it
Tommaso Caselli, Ahmet Üstün
2018 — conf
Characters or Morphemes: How to Represent Words?
Rep4NLP@ACL
Ahmet Üstün, Murathan Kurfali, Burcu Can
2017 J jnl
A Trie-Structured Bayesian Model for Unsupervised Morphological Segmentation.
CoRR
Murathan Kurfali, Ahmet Üstün, Burcu Can
2017 — conf
A Trie-structured Bayesian Model for Unsupervised Morphological Segmentation.
CICLing (1)
Murathan Kurfali, Ahmet Üstün, Burcu Can
2017 J jnl
Turkish PoS Tagging by Reducing Sparsity with Morpheme Tags in Small Datasets.
CoRR
Burcu Can, Ahmet Üstün, Murathan Kurfali
2016 — conf
Turkish PoS Tagging by Reducing Sparsity with Morpheme Tags in Small Datasets.
CICLing (1)
Burcu Can, Ahmet Üstün, Murathan Kurfali
2016 — conf
Unsupervised Morphological Segmentation Using Neural Word Embeddings.
SLSP
Ahmet Üstün, Burcu Can