📏 CUNI and Phrase at WMT25 MT Evaluation Task
Miroslav Hrabal, Ondrej Glembek, Aleš Tamchyna, Almut Silja Hildebrand, Alan Eckhard, Miroslav Štola, Sergio Penkale, Zuzana Šimečková, Ondřej Bojar, Alon Lavie, Craig Stewart
aclanthology.org/2025.wmt-1.68

🇨🇿 CUNI at WMT25 General Translation Task
Josef Jon, Miroslav Hrabal, Martin Popel, Ondřej Bojar
aclanthology.org/2025.wmt-1.44
Our submission to the WMT25 translation shared task showcases CUNI's latest approaches to general-purpose machine translation across multiple language pairs.

🔤 Pretraining Language Models with LoRA and Artificial Languages
Nalin Kumar, Mateusz Lango, @tuetschek.bsky.social t
aclanthology.org/2025.babylm-...
Constructed artificial languages with LoRA affects language model development.

🎓 You are an LLM teaching a smaller model everything you know: Multi-task pretraining of language models with LLM-designed study plans
Wiktor Kamzela, Mateusz Lango, @tuetschek.bsky.social
aclanthology.org/2025.babylm-...

🌍 Towards Adding Arabic to CorefUD
Dima Taji and Dan Zeman
aclanthology.org/2025.crac-1.6
Expanding the CorefUD universal coreference dataset to Arabic - taking important steps toward truly multilingual coreference resolution resources and better Arabic NLP.

📊 Real-World Summarization: When Evaluation Reaches Its Limits
@patuchen.bsky.social , @tuetschek.bsky.social , @saad.me.uk
aclanthology.org/2025.finding...
For hotel highlights, metrics like word overlap surprisingly match human judgments better than complex methods. LLMs unreliable as evaluators.

👥 Can Large Language Models Personalize Dialogues to Generational Styles?
P. Balestrucci, @tuetschek.bsky.social, L. Anselma, A. Mazzei
aclanthology.org/2025.finding...
Can LLMs adapt dialogues to generational styles? We show with P-MultiWoZ that models capture patterns from Boomers to Gen Z.

🤖 LLM Agents Implement an NLG System from Scratch
Mateusz Lango, Ondrej Dusek
aclanthology.org/2025.emnlp-i...
LLM agents can autonomously build interpretable, rule-based RDF-to-text generators from scratch, combining the LLMs with the transparency and reliability of traditional rule-based systems.

📚 SRS-Stories: Vocabulary-constrained multilingual story generation for language learning
Wiktor Kamzela, Mateusz Lango & @toonietuesday.bsky.social
aclanthology.org/2025.emnlp-i...
LLM stories teach vocab while reviewing learned words via Spaced Repetition-more grammatical than standard generation

The tools developed by Pavel and Jan are used by several oral history collections around the world, including the USC Holocaust testimonies available and the Center for Visual History Malach at UFAL. Their general ASR tools are also available at lindat.mff.cuni.cz/services/uwe....

Letní školu vedli: Ondřej Bojar, Zdeněk Kasner, Tomáš Polák, Dominik Macháček, Miroslav Hrabal a Josef Jon.

In his speech titled "Indirect Objects across Languages: A Trap in Universal Dependencies?" he discussed the challenges of the UD framework in relation to traditional language descriptions. He highlighted the ambiguities in UD guidelines and their impact on annotation practices.

ufal.mff.cuni.cz/grants/human...
Projekt bude zkoumat, jak mohou velké jazykové modely a další technologie umělé inteligence přispět k demokratickému dialogu, vzdělávání a komunikaci mezi lidmi.

Období realizace: 1. 3.2025 – 31. 12.2028. Financování: OP JAK: Společenské a humanitní vědy 2/2