Zobrazeno 1 - 10
of 197
pro vyhledávání: '"Koehn, Philipp"'
Large language models (LLMs) have achieved remarkable success across various NLP tasks, yet their focus has predominantly been on English due to English-centric pre-training and limited multilingual data. While some multilingual LLMs claim to support
Externí odkaz:
http://arxiv.org/abs/2410.03115
Autor:
Kocmi, Tom, Avramidis, Eleftherios, Bawden, Rachel, Bojar, Ondrej, Dvorkovich, Anton, Federmann, Christian, Fishel, Mark, Freitag, Markus, Gowda, Thamme, Grundkiewicz, Roman, Haddow, Barry, Karpinska, Marzena, Koehn, Philipp, Marie, Benjamin, Murray, Kenton, Nagata, Masaaki, Popel, Martin, Popovic, Maja, Shmatova, Mariya, Steingrímsson, Steinþór, Zouhar, Vilém
This is the preliminary ranking of WMT24 General MT systems based on automatic metrics. The official ranking will be a human evaluation, which is superior to the automatic ranking and supersedes it. The purpose of this report is not to interpret any
Externí odkaz:
http://arxiv.org/abs/2407.19884
Autor:
Lu, Taiming, Koehn, Philipp
This paper investigates the propagation of harmful information in multilingual large language models (LLMs) and evaluates the efficacy of various unlearning methods. We demonstrate that fake information, regardless of the language it is in, once intr
Externí odkaz:
http://arxiv.org/abs/2406.13748
Data availability limits the scope of any given task. In machine translation, historical models were incapable of handling longer contexts, so the lack of document-level datasets was less noticeable. Now, despite the emergence of long-sequence method
Externí odkaz:
http://arxiv.org/abs/2406.03869
Autor:
Wu, John F., Hyk, Alina, McCormick, Kiera, Ye, Christine, Astarita, Simone, Baral, Elina, Ciuca, Jo, Cranney, Jesse, Field, Anjalie, Iyer, Kartheik, Koehn, Philipp, Kotler, Jenn, Kruk, Sandor, Ntampaka, Michelle, O'Neill, Charles, Peek, Joshua E. G., Sharma, Sanjib, Yunus, Mikaeel
Large Language Models (LLMs) are shifting how scientific research is done. It is imperative to understand how researchers interact with these models and how scientific sub-communities like astronomy might benefit from them. However, there is currentl
Externí odkaz:
http://arxiv.org/abs/2405.20389
Non-autoregressive Transformers (NATs) are recently applied in direct speech-to-speech translation systems, which convert speech across different languages without intermediate text data. Although NATs generate high-quality outputs and offer faster i
Externí odkaz:
http://arxiv.org/abs/2405.13274
While Transformer-based neural machine translation (NMT) is very effective in high-resource settings, many languages lack the necessary large parallel corpora to benefit from it. In the context of low-resource (LR) MT between two closely-related lang
Externí odkaz:
http://arxiv.org/abs/2403.10963
Autor:
Tan, Weiting, Chen, Yunmo, Chen, Tongfei, Qin, Guanghui, Xu, Haoran, Zhang, Heidi C., Van Durme, Benjamin, Koehn, Philipp
We introduce STAR (Stream Transduction with Anchor Representations), a novel Transformer-based model designed for efficient sequence-to-sequence transduction over streams. STAR dynamically segments input streams to create compressed anchor representa
Externí odkaz:
http://arxiv.org/abs/2402.01172
Autor:
Shen, Lingfeng, Tan, Weiting, Chen, Sihao, Chen, Yunmo, Zhang, Jingyu, Xu, Haoran, Zheng, Boyuan, Koehn, Philipp, Khashabi, Daniel
As the influence of large language models (LLMs) spans across global communities, their safety challenges in multilingual settings become paramount for alignment research. This paper examines the variations in safety challenges faced by LLMs across d
Externí odkaz:
http://arxiv.org/abs/2401.13136
Autor:
Wang, Longyue, Tu, Zhaopeng, Gu, Yan, Liu, Siyou, Yu, Dian, Ma, Qingsong, Lyu, Chenyang, Zhou, Liting, Liu, Chao-Hong, Ma, Yufeng, Chen, Weiyu, Graham, Yvette, Webber, Bonnie, Koehn, Philipp, Way, Andy, Yuan, Yulin, Shi, Shuming
Translating literary works has perennially stood as an elusive dream in machine translation (MT), a journey steeped in intricate challenges. To foster progress in this domain, we hold a new shared task at WMT 2023, the first edition of the Discourse-
Externí odkaz:
http://arxiv.org/abs/2311.03127