PROGRAMME

The tutorials, the workshops and the main conference will take place at the same building.
Each workshop will run in parallel with the other and with the tutorial sessions on 7 September 2026.
The registration for all events will take place on Floor 2.

View parallel sessions schedule

Lecturer: Dimitar Hristov (Institute for Bulgarian Language, BAS)

09:00 – 10:30 | Tutorial Session

10:30 – 11:00 | Coffee Break

11:00 – 12:30 | Tutorial Session

Lecturer: Ivan Derzhanski (Institute of Mathematics and Informatics, BAS)

13:30 – 15:00 | Tutorial Session

15:00 – 15:30 | Coffee Break

15:30 – 17:00 | Tutorial Session

Speaker: Prof. Ruslan Mitkov (Lancaster University, University of Alicante)

Speaker: Prof. Preslav Nakov (Mohamed bin Zayed University of Artificial Intelligence)

10:15 – 10:20 | Break

• 10:20 – 10:40 – Irina Temnikova, Devora Kotseva, Ruslana Margova, Todor Kiriakov, Yolina Petrova, Nevena Grigorova, Alexander Komarov, Bogomil Katanov, Boryana Kostadinova and Stefan Minkov
  PROPer-BG: A New Publicly Accessible Package of Resources for Propaganda Detection in Bulgarian
• 10:40 – 11:00 – Jordan Kralev
  Direct Projection of Visual Features into Linguistic Embeddings for Domain-Adapted Multimodal Learning

• 11:30 – 11:50 – Svetla Boytcheva, Sylvia Vassileva and Pavel Boytchev
  BGSynthMedic: Synthetic Parallel English-Bulgarian Discharge Summary Corpus with Automatic Annotations
• 11:50 – 12:10 – George Pashev, Silvia Gaftandzhieva
  SymLang: A Structured Symbolic Intermediate Representation for Optimized Human-AI Communication in Assistive Clinical Technologies
• 12:10 – 12:30 – Yolina Petrova and Nikola Blajev
  Zero-Shot Detection of AI-Generated Bulgarian Text via Binoculars Framework

Chair: Hristina Kukova

• Mullosharaf Arabov
  A Systematic Benchmark of Machine Transliteration Models for the Tajik-Farsi Language Pair: A Comparative Study from Rule-Based to Transformer Architectures
• Mullosharaf Arabov, Karomatullo Habibullozoda and Nurali Shirinov
  TajikNLP: An Open-Source Toolkit for Comprehensive Text Processing of Tajik (Cyrillic Script)
• Öner Aytaş, Mehmet Ali Bayram, Tuğçe Şen, Banu Diri and Göksel Biricik
  Mitigating Security Vulnerabilities in Open-Source Large Language Models via a Supervisor Agent Architecture
• Abdelouahab Baflah
  A Qualitative Error Analysis of Whisper-Based Sentiment Recognition in Algerian Spoken Arabic: Prosodic and Pragmatic Failures in a Low-Resource Dialect
• Mehmet Ali Bayram, Öner Aytaş, Tuğçe Şen
  Morphology-Aware Tokenization Improves Turkish Sentence Embeddings under Controlled Training Conditions
• Nevena Grigorova, Stanislav Penkov, Keith Kiely, Dobromir Tsolyov, Iglika Ivanova
  Syntactic Patterns in Political Speech: Dependency-Based Ad Hominem Detection in Bulgarian Social Media
• Maria Khokhlova
  A Comparative Study of Synonym Generation in Russian with Instruction-Tuned Large Language Models in a Zero-Shot Setting
• Maria Khokhlova, Mikhail Koryshev
  Thematic Dominants of German-Language Catholic Hymns: Automatic Analysis using Word Embeddings
• Ruslana Margova
  Detecting Pseudoscientific Narratives in Bulgarian Science Communication: Nationalistic Mystification, Pseudo-Expertise and Cultural Blind Spots in AI-Assisted Analysis
• Mila Marcheva-Nash and Weiwei Sun
  PS-CHILDES-BG: A Constituency Treebank of Bulgarian Morphemically Tokenised Child-directed Speech
• Brandon Caillahua Mendoza
  Genetic Algorithms with Normalized Fitness Functions for Low-Resource Language Identification: A Case Study on Archaic Romanian and Medieval Slavonic Liturgical Manuscripts
• Junya Morita
  Corpus-based Study of Word Formation: Anti-agentive Nominals in English and Japanese

• 13:30 – 13:50 – Amina Laggoun, Youness Moukafih, Ouassim Karrakchou, Kamel Smaili
  Towards Frugal BERT Models for Maghrebi Dialects: Pruning and Distillation Approaches
• 13:50 – 14:10 – Ján Mačutek, Michaela Koščová, Michaela Nogolová and Radek Čech
  What to Take as the Basic Verb Form? A Pilot Study in Czech
• 14:10 – 14:30 – Radu Ion, Verginica Barbu Mititelu, Elena Irimia, Maria Mitrofan and Vasile Păiș
  Benchmarking Language Models on Romanian Eighth-Grade National Exams
• 14:30 – 14:50 – Silvie Cinková, Ivana Kvapilíková and Jan Černý
  Lexical Surprisal from Large Language Models: Cautionary Notes for Legal Discourse
• 14:50 – 15:10 – Natalia Dankova
  Analysing French Narrative Texts Generated by ChatGPT-5

• 15:30 – 15:50 – Verginica Barbu Mititelu, Elena Irimia, Simina Popa, Mihaela Cristescu, Catalin Mihaila, Lea Radu, Ioana Bucsa
  The Romanian Component of the Parallel ELEXIS Corpus: ELEXIS-Ro
• 15:50 – 16:10 – Atanas Atanasov
  A Dual-Interface Syntax Tree Editor with Automated Assessment for Learning Management Systems
• 16:10 – 16:30 – Maria Sampedro Mella
  Developing a Pragmatic Annotation System for a Spoken Corpus
• 16:30 – 16:50 – Tsvetelina Stefanova, Petya Osenova
  Evaluating the Linguistic Properties of AI-Generated Bulgarian News Texts
• 16:50 – 17:10 – Svetla Koeva
  Towards a New Grammar Checker for Bulgarian

Speaker: Dr Khalid Choukri (European Language Resources Association)

10:15 – 10:20 | Break

• 10:20 – 10:40 – Karla Csuros, Madalina Chitez, Roxana Rogobete and Petya Osenova
  Building a Cross-lingual Corpus of BA Theses: Rationale and Preliminary Observations
• 10:40 – 11:00 – Maria Todorova, Valentina Stefanova, Tsvetana Dimitrova, Mihaela Moskova, Hristina Kukova, Svetlozara Leseva, Ivelina Stoyanova, Dimitar Georgiev and Svetla Koeva
  Hybrid Benchmark for Long-Text Understanding and Reasoning in Bulgarian

• 11:30 – 11:50 – Marina Bagi, Aleksandra Marković and Ranka Stanković
  Toward a Serbian FrameNet: Resource Construction, Methodology, and Initial Annotation Results
• 11:50 – 12:10 – Stanislav Penkov
  A WordNet-Aligned Taxonomy of Logical Fallacies for Low-Resource NLP
• 12:10 – 12:30 – Ranka Stanković, Cvetana Krstev, Milica Ikonić Nešić and Miloš Utvić
  Relation Extraction from Text of Various Domains: A Serbian Case Study

• 13:30 – 13:50 – Andrea Nadalini, Claudia Marzi, Marcello Ferro, Vito Pirrelli, Maria Todorova, Valentina Stefanova, Tsvetana Dimitrova, Hristina Kukova and Svetla Koeva
  Assessing Reading Development through Finger-Voice Span: a Comparative Study of Bulgarian and Italian
• 13:50 – 14:10 – Ruslana Margova and Irina Temnikova
  Experiment Design of a Balanced Eye-Tracking Study for Measuring the Cognitive Impact of the Stimulus Word “Disinformation”
• 14:10 – 14:30 – Tatiana Sherstinova, Irina Petrova, Aleksei Melnik, Sofiia Chepovetskaia, Karina Azarevich, Viktoria Deborina
  Everyday Speech of Russian Students: From Field Recordings to the ESC Speech Corpus (Data Processing and Representation)
• 14:30 – 14:50 – Gunta Nešpore-Bērzkalne, Mikus Grasmanis and Ilze Lokmane
  Multi-word Expressions in the Electronic Dictionary Tēzaurs: Integrating Data from Different Sources
• 14:50 – 15:10 – Dobromir Tsolyov, Keith Kiely, Georgi Goranov, Iglika Ivanova, Stanimir Dimitrov
  Optimizing LLM-based Discourse Analysis in Low-Resource Languages: A Case Study of Bulgarian Euro Adoption Narratives

• 15:30 – 15:50 – Iglika Nikolova-Stoupak, Gaël Lejeune, Eva Schaeffer-Lacroix
  Quantifying Literary Style: A Corpus-Based Study
• 15:50 – 16:10 – Tomás Chismol Carbonell
  Zipfian Structure, Eurolect and Translationese: Exploratory Analysis of an EU Legal Corpus in English, French, Spanish and Bulgarian
• 16:10 – 16:30 – Iryna Karamysheva, Jakob Horsch
  Imploring You to Investigate More Languages: A Corpus-based Contrastive Study of Ukrainian and English Causative Constructions
• 16:30 – 16:50 – Ekaterina Tarpomanova
  Balkan Focus Particles in Bulgarian: bash and taman
• 16:50 – 17:10 – Ivan Derzhanski, Olena Siruk
  When Bulgarian ‘Still’ Meets Ukrainian ‘Already’

Lecturer: Irina Temnikova (Big Data for Smart Society Institute – GATE)

09:00 – 10:30 | Tutorial Session

10:30 – 11:00 | Coffee Break

11:00 – 12:30 | Tutorial Session

Lecturer: Ivan Ivanov (Stihia.ai)

13:30 – 14:30 | Tutorial Session

14:30 – 15:00 | Coffee Break

15:00 – 16:00 | Tutorial Session