Trina Dutta Barlow
Trina Dutta Barlow
Assistant Professor of Computer Science
- Doctor of Philosophy in Computer Science, Florida State University, Tallahassee, Fla., 2025
- Master of Science in Computer Science, Florida State University, Tallahassee, Fla., 2021
- Master of Science in Computer Science, St. Xavier's College, Kolkata, WB, India, 2019
- Bachelor of Science in Computer Science, St. Xavier's College, Kolkata, WB, India, 2017
During her doctoral studies, Trina Dutta Barlow worked on multiple projects, including the development of a pipeline for detecting and localizing hallucinations in abstractive summarization, a novel neuro-symbolic system aimed at enhancing logical reasoning in LLMs and a novel intervention technique designed to reduce hallucinations in abstractive summary generation. Her work has contributed to advancing the reliability of AI-generated content through fine-tuned models, semantic matching techniques, and intervention strategies.
Research Interests
- Deep Learning
- Natural Language Processing
- Object Recognition
- Neuro-symbolic Artificial Intelligence (AI)
- Logic Reasoning in Large Language Models
- Quantum Machine Learning
- AI Agents
Courses Taught
- Introduction to Computer Science with C++
- Python Programming
- Advance Python Programming
- Deep Learning Fundamentals
Certifications and Memberships
- International Joint Conference on Neural Networks
Selected Works
Publications
- "FaCTQA:Detecting and Localizing Factual Errors in Generated Summaries Through Question and Answering from Heterogeneous Models," 2024 International Joint Conference on Neural Networks (IJCNN), Yokohama, Japan, 2024, pp. 1-8, doi: 10.1109/IJCNN60899.2024.10650810, URL: https://ieeexplore.ieee.org/abstract/document/10650810
- "ILENS: Iterative Logical Enhancement via Neurosymbolic Computation and Common Sense", Submitted to ACL Rolling Reviews - June 2024, URL: https://openreview.net/forum?id=XJEmvFwrkT
