Institution profile

Jožef Stefan Institute

Academic institutioneurope · si
Official website
Research library107linked papers
Opportunities0open roles
Selected work

Representative Papers

Mi\'{c}i Princ -- A Little Boy Teaching Speech Technologies the Chakavian Dialect

Feb 03, 2026

This study addresses the scarcity of structured speech–text aligned data for the endangered Chakavian dialect, which has hindered its integration into artificial intelligence applications. We present the first high-quality, word-aligned multimodal dataset of *The Little Prince* in Chakavian, comprising synchronized text, images, and audio, meticulously curated through manual alignment and publicly released via the CLARIN.SI platform. Fine-tuning the Whisper-large-v3 model on this dataset yields substantial improvements in automatic speech recognition performance, reducing the word error rate by 50% and the character error rate by approximately two-thirds on the test set. This work establishes a reproducible data paradigm and technical framework for AI-driven preservation of endangered dialects.

1 citationsRead paper

Advancing the Scientific Method with Large Language Models: From Hypothesis to Discovery

May 22, 2025

Large language models (LLMs) face critical challenges in scientific research—including hallucination, low reliability, and ambiguous ethical accountability—hindering their trustworthy integration into the scientific process. Method: This paper repositions LLMs as “collaborative creative engines” and systematically investigates their deep integration across the full scientific workflow: hypothesis generation → experimental design → data analysis → discovery validation. We synergistically combine prompt engineering, scientific knowledge augmentation, verifiable reasoning-chain construction, and cross-disciplinary workflow integration. Contribution/Results: We introduce (1) the first comprehensive LLM application taxonomy spanning the entire scientific lifecycle; (2) a human-aligned, stage-specific evaluation framework with quantifiable collaboration metrics; and (3) an ethics governance mechanism balancing creative stimulation with responsibility constraints. Our work transcends the conventional instrumental use of LLMs, establishing both theoretical foundations and actionable pathways for AI-augmented scientific paradigm transformation.

1 citationsRead paper
Recent publications

Latest Papers