Tiberiu Sosea

I am a fourth-year Ph.D. student at the University of Illinois Chicago and I am advised by Professor Cornelia Caragea. My research interests lie in emotion detection and sentiment analysis. Some of my recent work focuses on developing new semi-supervised learning methods for emotion detection. I was an intern at Google with the Cloud AI and Counter Abuse Research teams. I did my undergrad in Romania at the Politehnica University of Bucharest.


  • Machine Learning
  • Computational Linguistics
  • Semi-supervised Learning
  • Computer Vision
  • Ph.D. in Natural Language Processing

    University of Illinois Chicago, 2019 - Present

  • BSc

    Politehnica University of Bucharest, 2015 - 2019


Google LLC - Google Cloud AI Modeling
Ph.D. Software Engineer Intern
Google LLC - Google Cloud AI Modeling
Jun 2021 – Aug 2021 Seattle, Washington
Created a new multitasking architecture for a document sequence splitting and classification task. The proposed method improves the flexiblity, is pre-trainable, and is more scalable than the previous approach.
Google LLC - Counter Abuse Research
Ph.D. Software Engineer Intern
Google LLC - Counter Abuse Research
Jun 2020 – Sep 2020 Mountain View, California
Created a low-level PDF parser, feature extractor, feature exporter and a machine learning pipeline aimed at identifying malicious PDF documents.
University of Illinois Chicago
Research Assistant
University of Illinois Chicago
Jan 2020 – Present Chicago, Illinois
Research in emotion detection and semi-supervised learning applied to text classification.


Tiberiu Sosea and Cornelia Caragea: eMLM: A New Pre-training Objective for Emotion Related Tasks   in ACL 2021

Tiberiu Sosea and Cornelia Caragea: CancerEmo: A Dataset for Fine-Grained Emotion Detection   in EMNLP 2020

Tiberiu Sosea and Cornelia Caragea: Leveraging Training Dynamics and Self-Training for Text Classification   in EMNLP Findings 2022

Hongli Zhan^, Tiberiu Sosea^, Cornelia Caragea, Junyi Jessy Li and Cornelia Caragea: Why Do You Feel This Way? Summarizing Triggers of Emotions in Social Media Posts   in EMNLP 2022

Tiberiu Sosea, Chau Pham, Alexander Tekle, Cornelia Caragea and Junyi Jessy Li: Emotion Analysis and Detection during COVID-19   in LREC 2022

Tiberiu Sosea and Cornelia Caragea: Ensynet: A Dataset for Encouragement and Sympathy Detection   in LREC 2022

Iustin Sirbu, Tiberiu Sosea, Cornelia Caragea, Doina Caragea, Traian Rebedea Multimodal Semi-supervised Learning for Disaster Tweet Classification   in COLING 2022

George Maratos, Tiberiu Sosea, Cornelia Caragea Label Smoothing for Emotion Detection   in AAAI 2023 Student Poster Program

Tiberiu Sosea, Iustin Sirbu, Cornelia Caragea, Doina Caragea and Traian Rebedea: Using the Image-Text Relationship to Improve Multimodal Disaster Tweet Classification   in ISCRAM 2021

Yingjie Li, Tiberiu Sosea, Aditya Sawant, Ajith Jayaraman Nair, Diana Inkpen and Cornelia Caragea: P-Stance: A Large Dataset for Stance Detection in Political Domain   in ACL Findings 2021

^ –> Equal Contribution