Large Language Models (LLMs) are increasingly used to generate synthetic textual data for training smaller specialized models.However, a comparison of various generation strategies for low-resource language settings is lacking.While various prompting strategies have been proposed-such as demonstrations, labelbased summaries, and self-revision-their comparative effectiveness remains unclear, especially for low-resource languages.In this paper, we systematically evaluate the performance of these generation strategies and their combinations across 11 typologically diverse languages, including several extremely low-resource ones.Using three NLP tasks and four open-source LLMs, we assess downstream model performance on generated versus gold-standard data.Our results show that strategic combinations of generation methods-particularly targetlanguage demonstrations with LLM-based revisions-yield strong performance, narrowing the gap with real data to as little as 5% in some settings.We also find that smart prompting techniques can reduce the advantage of larger LLMs, highlighting efficient generation strategies for synthetic data generation in lowresource scenarios with smaller models.

Jan Cegin, Jakub Simko, Peter Brusilovsky. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Jan Cegin, Branislav Pecher, Jakub Simko, Ivan Srba, Maria Bielikova, Peter Brusilovsky. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

The emergence of generative large language models (LLMs) raises the question: what will be its impact on crowdsourcing?Traditionally, crowdsourcing has been used for acquiring solutions to a wide variety of human-intelligence tasks, including ones involving text generation, modification or evaluation.For some of these tasks, models like ChatGPT can potentially substitute human workers.In this study, we investigate whether this is the case for the task of paraphrase generation for intent classification.We apply data collection methodology of an existing crowdsourcing study (similar scale, prompts and seed data) using ChatGPT and Falcon-40B.We show that ChatGPT-created paraphrases are more diverse and lead to at least as robust models.

Unit testing focused on Modified Condition/Decision Coverage (MC/DC) criterion is essential in development safety-critical systems. However, design of test data that meets the MC/DC criterion currently needs detailed manual analysis of branching conditions in units under test by test engineers. Multiple state-of-art approaches exist with proven usage even in industrial projects. However, these approaches have multiple shortcomings, one of them being the Path explosion problem which has not been fully solved yet. Machine learning methods as meta-heuristic approximations can model behaviour of programs that are hard to test using traditional approaches, where the Path explosion problem does occur and thus could solve the limitations of the current state-of-art approaches. I believe, motivated by an ongoing collaboration with an industrial partner, that the machine learning methods could be combined with existing approaches to produce an approach suitable for testing of safety-critical projects.