Erik Wang
HomeResearchTeaching

Selected publications

I'm interested in benchmarks, the science of deep learning, and AI for science.

–
HorizonMath: Measuring AI Progress Toward Mathematical Discovery with Automatic Verification

Preprint

–
HARDMath2: A Benchmark for Applied Mathematics Built by Students as Part of a Graduate Class

NeurIPS 2025

–
HARDMath: A Benchmark Dataset for Challenging Problems in Applied Mathematics

ICLR 2025

–
Humanity's Last Exam: A Benchmark of Expert-Level Academic Questions to Assess AI Capabilities

Nature

–
A Comparative Analysis of Tornadic Debris and Debris Fallout Signatures from the 10 - 11 December 2021 and 28 - 29 May 2019 Tornadoes

41st Conference on Environmental Information Processing Technologies

© 2026 Erik Wang