{"paper":{"title":"HARDMath2: A Benchmark for Applied Mathematics Built by Students as Part of a Graduate Class","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ahmed Roman, Alexander Dazhen Cai, Andrea Elizabeth Biju, Anne Mykland, Anton Marius Graf, Arjun Nageswaran, Ava Williamson, David Brown, David Ettel, Duy Thuc Nguyen, Egemen Bostan, Ekaterina S. Ivshina, Erik Y. Wang, Felix Yeung, Hongyu Liao, Isabella A. Silveira, James V. Roggeveen, Jorge Garc\\'ia Ponce, Kacper K. Migacz, Kelly Werker Smith, Kennan Gumbs, Lavon Sykes, Logan G. Wilson, Lucas Martin, Lucy S. Nathwani, Luke Zhu, Mai Hoang, Marc L. Descoteaux, Matthew Scott Tan, Michael P. Brenner, Miguel Fernandez, Minjae Kim, Nickholas Gutierrez, Peter Donets, Raglan Ward, Russell Yang, Shida Liu, Siddharth Dandavate, Will Flintoft, William Fredenberg, Yijun Wang, Yiming Chen, Yixing Tang, Yuyang Chen, Zhe He","submitted_at":"2025-05-17T00:52:49Z","abstract_excerpt":"Large language models (LLMs) have shown remarkable progress in mathematical problem-solving, but evaluation has largely focused on problems that have exact analytical solutions or involve formal proofs, often overlooking approximation-based problems ubiquitous in applied science and engineering. To fill this gap, we build on prior work and present HARDMath2, a dataset of 211 original problems covering the core topics in an introductory graduate applied math class, including boundary-layer analysis, WKB methods, asymptotic solutions of nonlinear partial differential equations, and the asymptoti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11774","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.11774/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}