Bestelakoa

Tesi-lanetan egiten diren datu-bilketak eta datu horiek aztertzeko sortzen diren tresnak sakabanatuta egoten dira unibertsitateetako biltegietan. Dispertsio horrek oztopatu egiten du datu horien edo tresnen berrerabilera. CLARIAH-EUS bezalako ikerketa-azpiegitura digitalen helburuetako bat da tesietan jasotako datuak erraz bilatzeko tresnak sortzea eta beste ikertzaileentzat ikusgarri eta berrerabilgarri jartzea. Esaterako, CLARIN-ERIC azpiegiturak, Virtual Language Observatory (VLO) tresnarekin, biltegi ezberdinetan dauden datuak eta tresnak bilatzeko

Ikasleek ikasketa-maila bakoitzean izan beharreko laburpen-gaitasuna ondo zehaztuta izatea oso garrantzitsua da irakasleak ikasleak ulertu duena sakontasunez jakiteko eta feedback egokia emateko; tamalez, beharrizan horri ez zaio behar bezala erantzuten. Horri buelta emateko, laburpenak lantzeko proposamenak egiteaz gain, nahitaezkoa da proposamen horiekin ikasleek nola jardun duten ondo aztertzea eta emaitzak biltzea, horrek emango baitigu ikasleek ikasketa-maila bakoitzean laburpenak biltzen dituen trebetasunetan izan beharreko maila zehazteko bidea.

This paper explores the complementary roles of bottom-up and top-down approaches in Basque terminology planning. It traces the evolution of language normalization efforts in Basque, emphasizing the institutional contributions of Euskaltzaindia, UZEI, and the Terminology Commission. While systematic terminology work methods are essential for expanding the use of Basque in new domains, corpus-driven descriptive approaches provide valuable insights into real term usage.

The proliferation of AI technologies has reinforced the importance of developing critical thinking skills. We propose leveraging Large Language Models (LLMs) to facilitate the generation of critical questions: inquiries designed to identify fallacious or inadequately constructed arguments. This paper presents an overview of the first shared task on Critical Questions Generation (CQs-Gen). Thirteen teams investigated various methodologies for generating questions that critically assess arguments within the provided texts.

We introduce a professionally translated extension of the TruthfulQA benchmark designed to evaluate truthfulness in Basque, Catalan, Galician, and Spanish. Truthfulness evaluations of large language models (LLMs) have primarily been focused on English. However, the ability of LLMs to maintain truthfulness across languages remains under-explored. Our study evaluates 12 state-of-the-art open LLMs, comparing base and instruction-tuned models using human evaluation, multiple-choice metrics, and LLM-as-a-Judge scoring.