Ghazal Khalighinejad
About Me
I am a PhD candidate in the Department of Computer Science at Duke University, fortunate to be advised by Prof. Bhuwan Dhingra, and previously advised by Sam Wiseman. Previously, I was an undergrad at Sharif University of Technology.
I am interested in advancing the capabilities of transformer-based models, with a focus on addressing problems in sciences. My research explores how neural architectures shape model performance, methods for embedding multimodal data, and strategies for handling challenges associated with long-context tasks.
News
- Spring 2025: “Training Neural Networks as Recognizers of Formal Languages” accepted at ICLR 2025!
- Spring 2025: MatViX has been accepted to NAACL 2025 main conference.
- Fall 2024: Check out our new preprint, MatViX: https://matvix-bench.github.io/, a benchmark for multimodal information extraction from long documents.
- Summer 2024: IsoBench has been accepted to the Conference on Language Modeling.
- Summer 2024: I started my internship at Adobe!
- Spring 2024: Our paper PNCExtract got accepted into Findings of ACL 2024.
- Spring 2024: We introduce IsoBench, a benchmark for multimodal models with isomorphic inputs.
- Fall 2023: Our paper Approximating CKY with Transformers got accepted into Findings of EMNLP 2023.
- Summer 2023: Got a full-tuition scholarship and funding from AI for Understanding and Designing Materials program for research in the intersection of NLP and Materials Science.
- Fall 2022: Attended NeurIPS 2022 in New Orleans.
- Spring 2022: My undergrad research Paper got accepted to ICALP 2022.
- Fall 2021: Started PhD at Duke.
Selected Publications
-
NAACL
To appear at NAACL 2025 -
COLM
Conference on Language Modeling -
ACL
-
EMNLP
-
ICALP
International Colloquium on Automata, Languages, and Programming (ICALP), 2022.
Services
Conference Reviewers: ACL Rolling Review
Powered by Jekyll and Minimal Light theme.