PNNL

Ian Stewart is a data scientist at Pacific Northwest National Laboratory (PNNL) who is developing methods to make AI models more transparent and more reliable in a variety of mission-specific applications. At PNNL, Stewart’s work involves training and deploying AI models to address a range of practical problems, including model validation and scientific data analysis. Stewart has worked on the Knowledge Marshaling project to develop user tools to validate AI output, and he has developed novel domain-specific foundation models as part of the Steel Thread project. In more practical applications, Stewart has contributed to PermitAI to help speed environmental permitting procedures with AI-powered search tools. His work has been published in various AI conferences and journals, including Neurocomputing and Knowledge Discovery and Data Mining. 

• Natural language processing
• AI interpretability
• Human-centered machine learning
• Linguistics

• PhD in Human-Centered Computing, Georgia Institute of Technology
• AB in Linguistics, Dartmouth College

• Reviewer for ACL, IEEE, AAAI
• Reviewer for NSF SBIR grants

• Outstanding Performance Award (2023, 2025)

2026

• Munikoti, S., I. Stewart, S. Horawalavithana, H. Kvinge, T. Emerson, S. Thompson. 2026. “Generalist Multimodal AI: A Review of Architectures, Challenges and Opportunities.” Neurocomputing (forthcoming). 

2025

• Meyur, R., H. Phan, K. Hayashi, I. Stewart, S. Sharma, S. Chaturvedi, M. Parker, D. Nally, S. Montgomery, K. Pazdernik, A. Jannesari, M. Halappanavar, S. Munikoti, S. Horawalavithana, A. Acharya. 2025. “Benchmarking LLMs for Environmental Review and Permitting.” KDD Workshop on Large Language Models for Scientific and Societal Advances. eprint 2407.07321, url https://arxiv.org/abs/2407.07321.

2024

• Horawalavithana, S., S. Munikoti, I. Stewart, H. Kvinge, K. Pazdernik. 2024. “SCITUNE: Aligning Large Language Models with Human-Curated Scientific Multimodal Instructions.” ACL Workshop on NLP for Science, Miami, Florida.
• Stewart, I. and R. Mihalcea. 2024. “Whose wife is it anyway? Assessing bias against same-gender relationships in machine translation.” Conference: Proceedings of the 5th Workshop on Gender Bias in Natural Language Processing (GeBNLP). DOI:10.18653/v1/2024.gebnlp-1.23.