Hugging Face
The PortNLP lab is led by Dr. Ameeta Agrawal.
At PortNLP, we explore new ways to advance language technologies -- from understanding complex texts to generating multilingual summaries. We aim to design systems that are effective, efficient, fair, and human-centered. We envision a future where language technologies are trustworthy and useful in everyday life.
PortNLP started in 2021 with PI and two students focusing on a few specific domains of text summarization and multilingual language models. As we have continued to grow, our past students and present members are contributing to a wide range of topics in natural language understanding and processing.
We are also affiliated with the Coco Lab, a sister research group focused on the intersection of computer science, management, and anthropology.
Research Themes
Language Understanding
- Long Context
- Ambiguity
- Contextual Grounding
Language Generation
- Summarization
- Multilingual Modeling
- Efficient NLP
Human-Centered NLP
- Affective Modeling
- Conversational Systems
- Bias Mitigation
Recent News
January 2026
- Accepted at the Findings of EACL 2026: SymCode: A Neurosymbolic Approach to Mathematical Reasoning via Verifiable Code Generation, Sina Bagheri Nezhad, Yao Li, Ameeta Agrawal
- Accetped at AIMedHealth (AAAI 2026): From Policy to Logic for Efficient and Interpretable Coverage Assessment, Rhitabrat Pokharel, Hamid Reza Hassanzadeh, Ameeta Agrawal
November 2025
- EMNLP 2025: Amber Shore presented Correct-Detect: Balancing Performance and Ambiguity Through the Lens of Coreference Resolution in LLMs.
- EMNLP 2025: Francielle Vargas and Jackson Trager presented our collaborative work with Dr. Ameeta Agrawal on multi-hop hate speech detection and explanation - MFTCXplain: A Multilingual Benchmark Dataset for Evaluating the Moral Reasoning of LLMs through Hate Speech Multi-hop Explanation.
- ICDM 2025: Yufei Tao will present Lost-in-the-Later: Framework for Quantifying Contextual Grounding in Large Language Models at the Workshop on Grounding Documents with Reasoning, Agents, Retrieval, and Attribution @ ICDM 2025.
- AAAI 2025 Fall Symposium Series: (Nov 6–8, Arlington VA) Our paper Towards Personalized Explanations for Health Simulations: A Mixed-Methods Framework for Stakeholder-Centric Summarization by Philippe Giabbanelli and Ameeta Agrawal will be presented.
October 2025
- CAPO: Confidence Aware Preference Optimization Learning for Multilingual Preferences (Rhitabrat Pokharel, Yufei Tao, Ameeta Agrawal)
- MTQ-Eval: Multilingual Text Quality Evaluation for Language Models (Rhitabrat Pokharel, Ameeta Agrawal)
Sponsors
This lab is grateful for the funding support from the following grants and awards:
