Programme
The 10th Conference
Human Language Technologies - the Baltic Perspective
Riga, Latvia
October 6-7, 2022
| October 6 | ||
| 09.15-10.00 | Registration | |
| 10.00-10.20 |
Opening Welcoming address by Lana Frančeska Dreimane, Head of the Research Programme Unit; Ministry of Education Science
Welcoming address by conference organizers: Normunds Grūzītis (Institute of Mathematics and Computer Science, University of Latvia), Andrejs Vasiļjevs (Tilde) and Inguna Skadiņa (PC Chair) |
Room 702 |
| 10.20-11.10 |
Invited speech Marko Grobelnik (chair: Andrejs Vasiļjevs) |
AI and NLP. Some breaking news. |
| 11.10-11.30 | Coffee break | |
| 11.30-12.30 |
Session 1: Translation Technologies Chair: Eduard Barbu |
Room 702 |
| 11.30.-11.50 | Andre Tättar, Taido Purason, Hele-Andra Kuulmets, Agnes Luhtaru, Liisa Rätsep, Maali Tars, Mārcis Pinnis, Toms Bergmanis, Mark Fishel | Open and Competitive Multilingual Neural Machine Translation in Production |
| 11.50-12.10 | Toms Bergmanis, Mārcis Pinnis | From Zero to Production: Baltic-Ukrainian Machine Translation Systems to Aid Refugees |
| 12.10-12.30 | Maali Tars, Andre Tättar, Mark Fishel | Cross-lingual Transfer from Large Multilingual Translation Models to Unseen Under-resourced Languages |
| 12.30-14.00 | Lunch | |
| 14.00-15.20 |
Session 2: Speech Technologies Chair: Askars Salimbajevs |
Room 702 |
| 14.00-14.20 | Aivo Olev, Tanel Alumäe | Estonian Speech Recognition and Transcription Editing Service |
| 14.20-14.40 | Henry Härm, Tanel Alumäe | Abstractive Summarization of Broadcast News Stories for Estonian |
| 14.40-15.00 | Einar Meister, Lya Meister | Estonian Elderly Speech Corpus – Design, Collection and Preliminary Acoustic Analysis |
| 15.00-15.20 | Liisa Rätsep, Rasmus Lellep, Mark Fishel | Estonian Text-to-Speech Synthesis with Non-autoregressive Transformers |
| 15.20-16.00 |
Introduction to posters and demos Chair: Normunds Grūzītis |
|
| 16.00-16.20 | Coffee break | |
| 16.20-18.00 | Posters & Demos | Room: 701 |
| Eduard Barbu, Verginica Barbu Mititelu | Evaluating Computational Models of Similarity against a Human Rated Dataset | |
| Rinalds Vīksna, Inguna Skadiņa | Multilingual Transformers for Named Entity Recognition | |
| Roberts Darģis, Ilze Auziņa, Inga Kaija, Kristīne Levāne-Petrova, Kristīne Pokratniece | Corpus Based Self-Assessment Platform for Latvian Language Learners | |
| Jolanta Kovalevskaitė, Loïc Boizou, Ieva Bumbulienė, Erika Rimkutė, Jurgita Vaičenonienė | Approaches and Resources for Lithuanian Collocation Research | |
| Sigita Dereškevičiūtė, Asta Kazlauskienė | Prosodic Phrasing in Lithuanian: Preparatory Study | |
| Asta Kazlauskienė, Sigita Dereškevičiūtė | Observations on the Prosodic Marking of Narrow Focus in Lithuanian | |
| Jurgita Kapočiūtė-Dzikienė, Askars Salimbajevs | Comparison of Deep Learning Approaches for Lithuanian Sentiment Analysis | |
| Joris Kazlauskas | On the Secondary Rhythmical Stress in the Connecting Vowels of the Standard Lithuanian Compounds | |
| Justina Mandravickaitė, Milita Songailaitė, Veronika Gvozdovaitė, Danguolė Kalinauskaitė, Tomas Krilavičius | Towards Synthetic Social Media Data | |
| 19.00- | Conference Dinner | |
| October 7 | ||
| 9.00-10.30 |
Country presentations Chair: Einar Meister |
Room 702 |
| 9.00-9.30 | Andrius Utka | Recent Developments of Language Technologies in Lithuania. |
| 9.30-10.00 | Kadri Muishnek | Recent Developments of Language Technologies in Estonia |
| 10.00-10.30 | Inguna Skadiņa, Baiba Saulīte, Ilze Auziņa, Normunds Grūzītis, Andrejs Vasiļjevs, Raivis Skadiņš, Mārcis Pinnis | Latvian Language in the Digital Age: The Main Achievements in the Last Decade |
| 10.30-10.50 | Coffee break | |
| 10.50-11.30 |
Invited speech Magnus Sahlgren (chair: Normunds Grūzītis) |
GPT-SW3: building the first large generative language model for Swedish |
| 11.30.-12.30 |
Session 3: Natural Language Understanding Chair: Andrius Utka |
Room 702 |
| 11.30-11.50 | Hele-Andra Kuulmets, Andre Tättar, Mark Fishel | Estonian Language Understanding: a Case Study on the COPA Task |
| 11.50-12.10 | Maksym Del, Mark Fishel | Similarity of Sentence Representations in Multilingual LMs: Resolving Conflicting Literature and a Case Study of Baltic Languages |
| 12.10-12.30 | Pēteris Paikens, Laura Rituma, Lauma Pretkalniņa | Towards word sense disambiguation for Latvian |
| 12.30-14.00 | Lunch | |
| 14.00-14.40 |
Invited speech Jan Hajič (chair: Inguna Skadiņa) |
Multilingual Language Technology in the age of Artificial Intelligence and Deep Neural Networks |
| 14.40-15.40 |
Session 4: Written Language Processing Chair: Kadri Muischnek |
Room 702 |
| 14.40-15.00 | Heiki-Jaan Kaalep | Looking into Estonian Syllabification |
| 15.00-15.20 | Virginijus Dadurkevičius, Andrius Utka | Estimating the Amount of Lithuanian Text Indexed by Global Search Engines |
| 15.20-15.40 | Matīss Rikters, Sanita Reinsone | How Masterly Are People at Playing with Their Vocabulary? |
| 15.40-15.55 | Closing | Room 702 |
| 16.00-17.00 | Good-bye coffee |
