🎉 KORPUS TŘICETILETÝ: Nová publikace oddělení Českého Národního korpusu FF UK u příležitosti 30. výročí ČNK 🎉 🔗 www.nln.cz/knihy/korpus... ÚFAL se na knize podílel nejen editorsky, ale i třemi klíčovými příspěvky. #NLP hashtag#Lingvistika hashtag#UFAL hashtag#PDT30 hashtag#CorpusLinguistics

New work from our team on morphology-aware tokenization evaluation, accepted to EACL 2026 Findings.

If you think labeling text spans with LLMs is easy, you probably have not tried it yourself (we have! 🙃). Any method you can think of – be it tagging, matching, or indexing – has flaws. In our new preprint, we tested them all 💪We also proposed how to improve one of them. arxiv.org/abs/2601.16946

Ondrej Bojar on prompt dilution and the role of red team in deep learning models, processes essential for understanding why AI models sometimes bypass safety guardrails, and how to mprove their robustness. Watch here (in Czech): 🔗 www.ceskatelevize.cz/porady/10969...

🍻 We asked around the office what "UFAL" means to them. A mix of academic rigor, hard work, family atmosphere, a great cup of coffee, or a unique team. Whether we are discussing NLP over beer or collaborating on a multi-generational project, the spirit of UFAL is all about community. Happy 2026! 👇

OpenEuroLLM has been mentioned in Forbes magazine among the biggest achievements of Czech science this year. forbes.cz/tricko-proti...

Check out @zdenekkasner.bsky.social's Animated LLM! A visualization of the innards of language models for you and your class. 👨‍🏫

Kristýna Onderková reports a successful poster presentation on Table-to-Text Generation Evaluation (abstract: openreview.net/forum?id=CbD...) AT EurIPS 2025 WORKSHOP "AI for Tabular Data". Her co-authors O.Plátek, Z.Kasner and O.Dusek share the success, but did not taste the special EurIPA beer!

Podrobně o workshopu k našemu společnému projektu Infoveillance na blogu Newton Media www.newtonmedia.cz/workshop-inf...

Na workshopu Infoveillance - prevence proti infodemiím budou představeny výsledky nejnovějšího sociologického šetření o používání sociálních médií a kvalitě informací v české populaci. Dozvíte se i novinky o využití technologií pro detekci anomálií v prostředí síťových digitálních médií.

🎧 Ondřej Bojar v novém díle podcastu Alma mater! 📺 Sledujte na YouTube (www.youtube.com/watch?v=I1vS...), poslouchejte na Spotify a v dalších podcastových aplikacích.

🎮✨ Nový projekt od absolventky @mff.unikarlova.cuni.cz spojuje hudbu, emoce a umělou inteligenci! Hra „Symphony of Adventure“ totiž místo obvyklých dotazníků sbírá data o emocích v hudbě hraním – a tím pomáhá trénovat AI.

EMNLP 2025 is over... and Milan Straka is bringing home an award! 🏆 CorPipe triumphed in the prestigious CRAC25 Shared Task, focusing on multilingual coreference resolution. Did Milan just CRACk it? We certainly think so! 😉 🔗 Find out more at arxiv.org/abs/2509.17858 #EMNLP2025 #CorPipe #CRAC25

If you speak/know speakers of Piedmontese or Neapolitan 🇮🇹, check out @gianlucavico.bsky.social's project, which collects crowd-sourced translations to study how LLMs handle these under-resourced dialects. Anyone can participate! 🎯

🗓️ Mark the dates!

The EU's 🇪🇺 HPLT project, coordinated by @ufal.mff.cuni.cz is at #EMNLP2025! It has supported it as a silver sponsor, disseminating HPLT results from our booth and through several papers. We'll continue to shape the future of multilingual datasets and models here and in @openeurollm.bsky.social!

Excited to share our work at #EMNLP2025! Our team is presenting 12 papers across the main conference and workshops, covering multilingual NLG, LLM agents, coreference resolution, and machine translation. A thread with highlights 🧵👇

With @andrei-a-manea.bsky.social, we posted a survey on multilingual vision-language models 👉 arxiv.org/pdf/2509.22123 We reviewed 31 models+21 benchmarks. There's a tension between language neutrality (same results across languages) & cultural awareness (context matters differently across cultures)

Zveme na dnešní přednášku Jazykovědného sdružení, kterou od 17:30 přednese prof. PhDr. Eva Hajičová, DrSc. 🔗 Můžete přijít osobně nebo sledovat na zoomu: lnkd.in/eQeST-uG Téma přednášky: Aktuální členění v době paralelních korpusů 📸 Foto: Vladimír Šigut, UK

🚀 PROJECT LAUNCH: Infoveillance is Live! Our AI tool monitors digital media to detect misinformation and enhance public trust/literacy. Fighting infodemics & polarization. [https://ufal.mff.cuni.cz/grants/infoveillance] #Infoveillance #AI #Misinformation #PublicTrust #UFAL

Huge win! 🎉 The CLARIN Steven Krauwer Prize 2025 goes to LINDAT/CLARIAH-CZ partners Pavel Ircing & Jan Švec (UWB). They won for their ASR tools for complex Oral History recordings. Their work is vital for global digital humanities! #CLARIN #LINDAT #StevenKrauwerAward #ASR #OralHistory #DG

Šest kolegů vedlo pro DGT (Evropské ředitelství pro překlady) třídenní letní školu v Lucemburku. Učili 40+ pracovníků DGT nejnovější metody strojové podpory překladu a zajištění kvality. Cíl? Zefektivnit překlad legislativy EU do všech členských jazyků! #DGT #UFAL #StrojovyPreklad #AI #EUTools

@Dan Zeman has been invited as a keynote speaker at the ICLC 11 conference! iclc11.ff.cuni.cz/keynote-spea... #UFAL #ICLC11 #UniversalDependencies #CharlesUniversity #Prague

Nahlédněte na kick‐off meeting projektu ✨HumanAId: AI zaměřená na člověka pro udržitelnou a adaptabilní společnost✨. Projekt se silnou účastí: vede ho FFUK ve spolupráci s MFF UK, FSV UK, PF UP v Olomouci, FÚ AV ČR, prg.ai a Kampusem Hybernská. #prgAI #HumanAId #OPJAK 1/2

And another successfully defended thesis: 👉Dr.👈 Kira Droganova defended her thesis: Dependency Parsing beyond Simple Trees, which focused on enriching syntactic parsing with deeper semantic layers to better capture meaning across languages. Congratulations 🥳

🎉 Congratulations to 👉Dr.👈 Tomáš Musil on successfully defending his PhD thesis! 🍻 His talk explored #LLMs, theories of meaning, and their role in LLM #interpretability, highlighting unsupervised discovery of binary semantic features via ICA and the word intruder test.

Workshop "Regulace, AI a advokacie – zákulisí legislativy a advokátních inovací" představil OpenEuroLLM jako naději pro evropskou digitální suverenitu a nutnost pro konkurenceschopnost Evropy. Jan Hajič zdůraznil, že Česko se snaží o snižování byrokracie v oblasti AI. #AI #AIregulation #FutureOfLaw

Researchers' Night with @informatfyz.cuni.cz! You can come to a live podcast recording and try out a real-time automatic interpreting system ELITR. The event is on September 26th. 🔗 czechia.representation.ec.europa.eu/evropsky-den... #ELITR #AI #Interpreting #MachineTranslation #LanguageTech

Great week at #TSD2025 in Erlangen! Our @ufal-cuni.bsky.social team presented 7 papers covering various topics from Czech speech assessment to multilingual morphology. Thanks to all attendees who engaged with our work! 🧵👇 #NLP #ComputationalLinguistics #CzechNLP #MachineLearning

Šimon Libřický defended his bachelor thesis with great success and co-authored an article accepted to @ISMIR 2025. His topic is the intersection between music and computers - a model of difficulty of saxophone scores. Inteview for @informatfyz.cuni.cz: www.mff.cuni.cz/en/public/ne.... #UFAL #MFFUK

Tomorrow, Thu Aug 28, 14-17, join AutoMin 2025 fully remote session to learn about automatic summarization of meetings. Keynote talks by our colleagues from Zoom and Translated.net Link+Schedule: ufal.github.io/automin-2025 Proudly co-organized by Ondřej Bojar and folks from @ufal-cuni.bsky.social

Our team of 10 is at #MTMarathon2025 in Helsinki 🇫🇮, a week-long meeting of machine translation researchers, developers. ✅ Posters presented ✅ Now working on cool collaborative projects with researchers from around the world. #MachineTranslation #NLP

Our latest contribution to multilingual NLP evaluation 🚀 by @jlibovicky.bsky.social, @jindrahelcl.bsky.social, @andrei-a-manea.bsky.social & @gianlucavico.bsky.social

We are hosting a summer school „Data Literacy with R for Students of Humanities“ at Malá Strana, August 4-15. ufal.mff.cuni.cz/events/summe... #DataLiteracy #Humanities #Matfyz #UFAL #CharlesUniversity

#ACL2025NLP continues with two days of workshop, and @ufal-cuni.bsky.social folks are there with more than a dozen posters 👇

#ACL2025NLP in Vienna 🇦🇹 starts today with 23 🤯 @ufal-cuni.bsky.social folks presenting their work both at the main conference and workshops. Check out our main conference papers today and on Wednesday 👇

🎉 Congratulations to @zdenekkasner.bsky.social for winning 2nd Prize in the Joseph Fourier Prize for Computer Science! The Joseph Fourier Prize recognizes outstanding scientific work in computer science and is awarded annually by the French Embassy in the Czech Republic. 🇨🇿 🇫🇷

CUNI system submission for IWSLT 2025 Simultaneous Shared Task using Whisper and EuroLLM achieves great quality with 2-4 seconds latency 🤯 and outperforms baseline by a large margin. It uses context, prompting, and in-context learning. It's among the top-performing systems 🎖️in IWSLT 2025.

Accepted long paper at IWSLT 2025! 🥳 📆 31 July-1 August 2025 📍 Vienna, Austria Prompting LLMs: Length Control for Isometric Machine Translation, by Dávid Javorský, Ondřej Bojar, @yvofr.bsky.social, in collaboration with @ufal-cuni.bsky.social .

Research on trustworthy text generation by @tuetschek.bsky.social was featured in the Czech edition of @wired.com 😊 www.wired.cz/clanky/vyvij...

The pre-print of Dávid's ACL paper here: arxiv.org/abs/2506.04848 🗣️ MockConf, a 7-hour student interpreting dataset in 5 languages + InterAlign, a tool for analyzing simultaneous interpretation to advance automatic real-time translation.

Participate in the Terminology shared task at WMT 2025! (And there are some other cool tasks too.)

Přečtěte si rozhovor s @janastrakova.bsky.social na matfyz.cz! www.matfyz.cz/clanky/holky...

We got 3 papers accepted to the #ACL2025 main conf: 👉 An Expanded Massive Multilingual Dataset for High-Performance Language Technologies arxiv.org/abs/2503.10267 by @hajicjan.bsky.social, @jindrahelcl.bsky.social and many others: Data of 8T tokens in 193 langs + 380M parallel sentences in 51 langs

#NAACL2025 ended more than a week ago & @ufal-cuni.bsky.social folks were there: Main conf: @kathaem.bsky.social presented joint work w/ @tomlim.bsky.social, @jlibovicky.bsky.social and Alex Fraser: Beyond Literal Token Overlap: Token Alignability for Multilinguality aclanthology.org/2025.naacl-s...

On Saturday, @patuchen.bsky.social talked at the GHOST Day: Applied Machine Learning Conference in Poznań 🇵🇱 👉 Evaluating LLM-generated text at a scale 👈 lessons learned from evaluating hotel highlights and a script of a theater play, and practical recommendations for LLMs in a referenceless scenario

Collegium Carolinum, the Bavarian Academy of Sciences & @ufal-cuni.bsky.social cordially invite you to their workshop 👉 Automated Context? Practices, Opportunities and Risks of AI-Driven Translation in Bohemistics and the Humanities 🔗 www.collegium-carolinum.de/veranstaltun... 📍 Prague, May 15-16

🎶 Umíš plynně česky, je ti více než 18 a znáš píseň „Není nutno“? 🗣️ Přidej se tedy k výzkumu o řeči a zpěvu! 📅 Kdy a kde? 17.–18. 6. 2025 v Praze! 📲 Naskenuj QR kód z obrázku a zapoj se do studie ještě dnes! ❓ Dotazy směřuj na: [email protected] / [email protected]

Got a tokenization paper that just didn't make the cut for ICML? Submit it to the Tokenization Workshop TokShop at #ICML2025 -- we'd love to see it there! tokenization-workshop.github.io

The 👉Machine Learning Prague 2025👈 is happening right now! Today, @patuchen.bsky.social and @navitas.bsky.social presented their posters on text generation with LLMs. Also, don't miss @tuetschek.bsky.social's invited talk tomorrow at 11 a.m.