Tilde AI Lab:
Research and development on multilingual AI
In-depth research expertise areas
Machine translation
We innovate in domain-adapted, adaptive, and term-aware neural machine translation, developing methods for robustness, mitigation of biases, and large language models.
Conversational AI
We focus on multilingual natural language understanding, semantic indexing, LLM-based retrieval-augmented generation, and personalisation.
Speech technologies
We research cascaded and end-to-end speech recognition and speech translation, multi-speaker and multilingual speech synthesis, real-time speech recognition, automatic subtitling and dubbing.
Text analysis
We develop multilingual named entity recognition, anonymisation and pseudonymisation, term recognition and extraction, morphological analysis, lemmatization, part-of-speech tagging and other tools.
Knowledge management
We maintain the largest termbank in Europe - EuroTermBank, and develop term management and electronic dictionary tools.
TildeOpen LLM now powers our machine translation
Current research projects
CoRDS is a the Marie SkÅodowska-Curie Actions (MSCA) Doctoral Networks programme project aiming to address gaps in AI -powered decision support tools by advancing the research on data-driven optimization methods and training a new generation of experts skilled in the combination of Operations Research (OR) and trustworthy Machine Learning (ML). CoRDS goes beyond the state-of-the-art data-driven optimization methods by proposing decision support frameworks that combine OR and ML methods and enable robust, transparent, and fair solutions that reflect user preferences and address complex, uncertain real-world scenarios.
TheĀ LLMs4EU projectĀ aims to preserve European linguistic and cultural diversity in the digital age through cooperation between economic and academic actors. The project brings together Europeās leading players in the field of generative AI to ensure that European companies and especially SMEs have access to the tools and resources to become competitive regarding language technologies and especially Large Language Models (LLMs). The project aims to make LLMs and all the tools necessary for their exploitation in all EU languages available in open data by capitalizing on existing European programs and competencies.
The LINGUARISE-DC project addresses the pressing need for advanced natural language processing in Europeās defence sector. International missions face challenges such as linguistic diversity, limited training time, dependence on human interpreters, and security risks. LINGUARISE-DC seeks to transform language capabilities with an advanced human language technology (HLT) platform, designed to enhance European Union-led coalition missions and special operations.
Latest publications
Raivis SkadiÅÅ”, Daiga Deksne, Andris Hohbergs, RÅ«dolfs Jaunzars, Andrejs Petrovs, JustÄ«ne RÅ«dule and MÄrcis Pinnis. 2026. Dataset and Baseline Implementation for Retrieval-Augmented Procurement Validation. Lecture Notes in Networks and Systems, vol. 1949. Springer.
Rinalds VÄ«ksna and MÄrcis Pinnis. 2026. Pseudonymisation for Morphologically Rich Languages. Proceedings of 29th International Conference on Text, Speech and Dialogue. Lecture Notes in Artificial Intelligence, Springer.
Daiga Deksne (Tilde), Raivis SkadiÅÅ” (Tilde), MÄrcis Pinnis (Tilde), Andris Hohbergs, RÅ«dolfs Jaunzars, Andrejs Petrovs and JustÄ«ne RÅ«dule. 2026. Retrieval-Augmented Generation for Procurement Validation. Proceedings of the 18th International Conference on Agents and Artificial Intelligence, 4048-4055.
Our research team
MÄrcis Pinnis
Dr sc. comp., Chief AI Officer
Toms Bergmanis
Dr M. Inf., Researcher
Martins Kronis
M. sc. comp., Researcher/Developer
Jurgita KapoÄiÅ«tÄ-DzikienÄ
Dr sc. comp., Researcher
Raivis SkadiÅÅ”
Dr sc. comp., CTO
Andrejs Vasiļjevs
Inguna SkadiÅa
Matīss Rikters
Daiga Deksne
Dr philol., Mg. sc. comp., Mg. psych., Software Architect
Rinalds VÄ«ksna
Mg. sc. comp., Researcher
DÄvis Nicmanis
Mg. sc. comp., Researcher/Developer
Roberts Rozis
BSc. comp., Data Group Manager