Mono-and multilingual GPT-3 models for Hungarian

ZG Yang, LJ Laki, T Váradi, G Prószéky - International Conference on Text …, 2023 - Springer
In recent years, the growth in size of Transformer-based language models has accelerated
significantly. Global technology companies are training larger and larger models that require …

Efficiently adapting pretrained language models to new languages

Z Csaki, P Pawakapan, U Thakker, Q Xu - arXiv preprint arXiv:2311.05741, 2023 - arxiv.org
Recent large language models (LLM) exhibit sub-optimal performance on low-resource
languages, as the training data of these models is usually dominated by English and other …

[PDF][PDF] Jönnek a nagyok! BERT-Large, GPT-2 és GPT-3 nyelvmodellek magyar nyelvre

KÁ Kinga, LL János, LN Noémi, V Noémi, V Tamás - 2023 - core.ac.uk
Kivonat Az utóbbi években rendkívüli mértékben felgyorsult a Transformer alapú
nyelvmodellek méretének a növekedése. A globális technológiai cégek nagyobbnál …

Winograd schemata and other datasets for anaphora resolution in Hungarian

N Vadász, N Ligeti-Nagy - Acta Linguistica Academica, 2022 - akjournals.com
Abstract The Winograd Schema Challenge (WSC, proposed by Levesque, Davis &
Morgenstern 2012) is considered to be the novel Turing Test to examine machine …

Improve Performance of Fine-tuning Language Models with Prompting

ZG Yang, N Ligeti-Nagy - INFOCOMMUNICATIONS JOURNAL, 2023 - real.mtak.hu
This paper explores the effectiveness of prompt programming in the fine-tuning process of a
Hungarian language model. The study builds on the prior success of prompt engineering in …

P4Query: Static analyser framework for P4

D Lukács, G Tóth, M Tejfel - Annales Mathematicae et Informaticae, 2023 - real.mtak.hu
There are many important tasks in a conventional software development process which can
be supported by different analysis techniques. P4 is a high level domain-specific language …

The First Instruct-Following Large Language Models for Hungarian

ZG Yang, R Dodé, G Ferenczi, P Hatvani… - 2024 IEEE 3rd …, 2024 - ieeexplore.ieee.org
In recent months, large language models have gained significant attention, with companies
striving to develop models capable of solving various natural language processing tasks …

BiVaSE: A bilingual variational sentence encoder with randomly initialized Transformer layers

B Nyéki - Acta Linguistica Academica, 2022 - akjournals.com
Transformer-based NLP models have achieved state-of-the-art results in many NLP tasks
including text classification and text generation. However, the layers of these models do not …

Building machine reading comprehension model from scratch

ZG Yang, N Ligeti-Nagy - Annales Mathematicae et Informaticae, 2023 - real.mtak.hu
In this paper, we introduce a machine reading comprehension model and how we built this
model from scratch. Reading comprehension is a crucial requisite for artificial intelligence …

[PDF][PDF] HuBERTUSz: Alacsony paraméterszámú transzformer modellek létrehozása és kiértékelése magyar nyelvre

F Tamás, B Gábor - inf.u-szeged.hu
Kivonat Hazánkban is megnőtt az érdeklődés a transzformer modellek alkalmazása iránt.
Éppen ezért a modern előtanítási standardoknak megfelelően (pl. dinamikus maszkolás …