PHD Student in Natural Language Processing HiTZ Basque Center for Language Technologies Ixa NLP Group University of the Basque Country UPV/EHU Donostia-San Sebastián iker.garciaf@ehu.eus

About Me

I am currently a PhD candidate in Natural Language Processing (NLP) at the University of the Basque Country UPV/EHU, IXA Group, and HiTZ Basque Center for Language Technologies, funded by a grant from the Basque Government. My advisors are German Rigau and Rodrigo Agerri. I anticipate completing my PhD by April 2024.

My previous experiences include an internship as an Applied Scientist at Amazon Barcelona, where I was part of Lluis Marquez's team. I also served as Visiting Associate for 4 months at the School of Engineering and Applied Science, Department of Computer and Information Science, Cognitive Computation Group at the University of Pennsylvania under the supervision of Dan Roth.

My research primarily focuses on Multilingual Natural Language Processing. I aim to develop deep learning models and resources that enable NLP in languages with limited or no available resources. This research branches in two directions. First, data-transfer methods for which I have developed state-of-the-art techniques to automatically generate annotated data for languages that lack these resources. Second, model-transfer methods, a field in which I've made significant contributions to improve the zero-shot cross-lingual performance of NLP models. Recently, my research has branched into training Large Language Models (LLMs) for various tasks and domains. The most notable ones include GoLLIE a 34B parameter LLM which achieves state-of-the-art results for zero-shot Information Extraction, and MedicalMT5, the first open-source text-to-text multilingual model for the medical domain. I am also involved in different efforts towards fair and unbiased evaluation of LLMs.

Social Profiles