معرفی
Tomer Wolfson is a Postdoctoral Research Fellow at the University of Pennsylvania and a member of the Cognitive Computation Group led by Prof. Dan Roth. He is also a CHE Postdoctoral Fellow for AI and Data Science, focusing on the intersection of Data Management and Natural Language Processing.
- PhD from Tel Aviv University (2019-2024), advised by Prof. Daniel Deutch and Prof. Jonathan Berant
- Research intern at Allen Institute for AI (2019-2024)
- BSc and MSc in computer science from Tel Aviv University
Wolfson's research focuses on developing algorithms for understanding and reasoning over complex questions spanning hundreds of documents. His work includes creating benchmarks like MoNaCo and QAMPARI, analyzing language model reasoning paths, and improving retrieval-augmented generation systems. Key contributions include:
- MoNaCo: A benchmark for multi-document reasoning
- QAMPARI: Open-domain QA benchmark with multi-answer evaluation
- Break It Down: Question understanding benchmark
- Weakly supervised Text-to-SQL parsing systems
His work combines natural language processing, machine learning, and database systems to enhance automated reasoning capabilities. Wolfson has received recognition for his contributions to retrieval-augmented language models and parametric knowledge analysis.
Scientific awards include:
- CHE Postdoctoral Fellow for AI and Data Science
Wolfson has served as a lecturer at Tel Aviv University for "Web Data Management & Information Retrieval" (2021-2022) and as a teaching assistant for "Web Data Management" (2018-2020). He is currently affiliated with the Cognitive Computation Group at the University of Pennsylvania.
