Open to opportunities

William Soto

@williamsoto

Message

PhD researcher specializing in multilingual AI and text generation.

France

Message

What I'm looking for

I seek a collaborative environment where I can leverage my AI expertise and contribute to innovative projects.

I am a dedicated PhD researcher at the Centre National de la Recherche Scientifique in France, where I focus on applying generative AI, particularly large language models, to multilingual text generation and evaluation. My work has led to multiple high-level publications in prestigious conferences such as ACL, IJCNLP, and INLG, showcasing my commitment to advancing the field of natural language processing.

Throughout my academic journey, I have developed open-source code and models that contribute to the multilingual generation landscape. My research experience spans various roles, including the development of a multilingual paraphrase evaluation metric and a language identification model. I hold a Master's degree in Natural Language Processing from Université de Lorraine, where I graduated with honors, and a Bachelor's degree in Computer Sciences from Universidad de Costa Rica.

Experience

Work history, roles, and key accomplishments

Current

PhD Researcher

Current

Centre National de la Recherche Scientifique

Oct 2021 - Present (3 years 8 months)

Conducted research for one of the largest fundamental science agencies in Europe, applying generative AI and LLMs to multilingual text generation and evaluation from Knowledge Graphs. Published multiple papers in ACL Anthology conferences and developed publicly available open-source code and models.

Generative AI LLMs)Knowledge Graphs Python PyTorch Transformers Open Source

M2 Researcher

Centre National de la Recherche Scientifique

Mar 2021 - Aug 2021 (5 months)

Developed a Multilingual Paraphrase Evaluation metric. This involved applying natural language processing techniques to assess the semantic similarity of text.

Multilingual Paraphrase Evaluation Natural Language Processing Python Scipy scikit learn

M1 Researcher

Centre National de la Recherche Scientifique

Jun 2020 - Jul 2020 (1 month)

Developed a Language Identification model. This project focused on applying machine learning to determine the language of a given text.

Language Identification Machine Learning Python Pandas

Research and Teaching Assistant

Universidad de Costa Rica

Mar 2017 - Aug 2019 (2 years 5 months)

Conducted research in Data Anonymization, Multi-Agent Simulation, and Cloud Architecture. Supported teaching activities related to these advanced computing topics.

Anonymization Cloud Architecture Research Teaching SQL

Education

Degrees, certifications, and relevant coursework

Université de Lorraine

PhD in Informatics, Informatics

Pursued a PhD in Informatics with a thesis focused on Multilingual Graph-to-Text Generation and Evaluation. The defense is pending for October 2025.

Université de Lorraine

MSc in Natural Language Processing, Natural Language Processing

Grade: 16.40/20.00

Completed a Master of Science in Natural Language Processing. The thesis, 'X-ParEval: A Multilingual Metric for Paraphrase Evaluation,' contributed to the field. Achieved a grade average of 16.40/20.00, ranking 4th out of 20 students, and received a 'Très Bien' mention.