# Aaron Steven White (@aaronstevenwhite.io)

Profile: https://sifa.id/p/aaronstevenwhite.io
Headline: Computational Semanticist

## About

I'm a researcher in computational linguistics and natural language processing, developing methods that bridge machine learning with language science to build more interpretable and capable AI systems.

My work focuses on three core areas: (1) creating large-scale semantic understanding systems that can extract and synthesize information across documents, (2) developing interpretable neural architectures for language understanding, and (3) building evaluation frameworks that reveal how well AI systems truly understand language.

I have a proven track record of developing state-of-the-art NLP systems, including work in event-keyed summarization, multilingual information extraction, and semantic parsing. My research combines rigorous linguistic theory with practical applications.

I've published extensively in top venues across computational linguistics (TACL, ACL, EMNLP), cognitive science (Cognitive Science, Cognitive Psychology), and linguistics (Semantics & Pragmatics, Language Acquisition, Glossa). I've released numerous open-source datasets and codebases that enable researchers to build better natural language understanding systems—including large-scale natural language inference datasets and benchmarks for evaluating cross-lingual information extraction and complex event understanding.

I have extensive experience managing large-scale research initiatives and multi-institutional collaborations. As PI on multiple federally-funded projects, I've successfully:

\* Led multi-year cross-institutional research programs involving teams of linguists, computer scientists, and data scientists.
\* Managed large-scale data collection efforts with 1000+ annotators, implementing quality control and active learning pipelines.
\* Mentored and managed teams of research scientists, postdoctoral fellows, and graduate and undergraduate researchers, providing hands-on training in data collection, computational modeling, and software development.
\* Integrated research outputs into educational curricula, developing new computational linguistics programs at both undergraduate and graduate levels.

I'm currently exploring neurosymbolic approaches to meaning representation, controllable reasoning and summarization, information retrieval with highly structure queries, and LLM interpretability. I'm always interested in collaborating with teams tackling challenging problems in language understanding, reasoning, information extraction and retrieval, and computational semantics.

## Experience

- **Assistant Professor at University of Rochester** (2017 – 2023)
- **Associate Professor at University of Rochester** (2023 – present)
- **Postdoctoral Fellow at Johns Hopkins University** (2015 – 2017)

## Education

- **University of California, Santa Cruz** — BA, Linguistics (2006 – 2009)
- **University of Maryland, College Park** — PhD, Linguistics (2010 – 2015)

## Publications

- Machine-readable attitudes — aaronstevenwhite.leaflet.pub (https://aaronstevenwhite.leaflet.pub/3miwsz2hdv22i)