Felipe Russi

Department of Mathematics, Department of Computer Science and Systems Engineering, Universidad de los Andes.

prof_pic.jpeg

I graduated with dual majors in Computer and Systems Engineering and Mathematics from Universidad de los Andes. My research focuses on statistical learning and optimization methods for understanding uncertainty, structured dependencies, and reliability in AI systems deployed in human-centered and socially impactful domains such as healthcare, accessibility, and human–computer interaction.

In Mathematics, I was advised by Mauricio Junca, working on Variational Inference and Optimal Transport. I continue collaborating with Rubén Manrique on Natural Language Processing applied to biomedical texts. I also worked with Angelique Taylor at Cornell Tech’s AIRLab, developing interfaces for tele-operating robots with the goal of using them in medical environments.

I am particularly interested in statistical learning, causal inference, and computational optimization methods that enable principled evaluation and improvement of AI systems operating in complex real-world human settings.

contact & more:

email (academic): af [dot] ariasr [at] uniandes [dot] edu [dot] co

email (personal): felipea2811 [at] gmail [dot] com

  • In my free times, I like to watch movies.
  • The pronunciation of Felipe is [feˈli.pe]/”fe-lee-pe”.
  • My last name is Arias-Russi.

news

Nov 01, 2025 Publication at EMNLP 2025, TSAR Workshop Shared Task Uniandes at TSAR 2025 Shared Task: Multi-Agent CEFR Text Simplification with Automated Quality Assessment and Iterative Refinement. First author. 4th place in the TSAR 2025 Shared Task. Published in Proceedings of the Fourth Workshop on Text Simplification, Accessibility and Readability (TSAR 2025) at EMNLP 2025, Suzhou, China.
Nov 01, 2025 Publication at EMNLP 2025, TSAR Workshop A Multi-Agent Framework with Diagnostic Feedback for Iterative Plain Language Summary Generation from Cochrane Medical Abstracts. First author. Published in Proceedings of the Fourth Workshop on Text Simplification, Accessibility and Readability (TSAR 2025) at EMNLP 2025, Suzhou, China.
Oct 20, 2025 Extended abstract at ICCV 2025 – CV4A11y Workshop (Vision Foundation Models and Generative AI for Accessibility) Guiding Multimodal Large Language Models with Blind and Low Vision Visual Questions for Proactive Visual Interpretations.
Co-author. Accepted as an extended abstract at the CV4A11y Workshop, part of ICCV 2025.
OpenReview link
May 04, 2025 Publication at NAACL 2025, CL4Health Workshop Bridging the Gap in Health Literacy: Harnessing the Power of Large Language Models to Generate Plain Language Summaries from Biomedical Texts. First author. Published in Proceedings of the Second Workshop on Patient-Oriented Language Processing (CL4Health) at NAACL 2025, Albuquerque, New Mexico.