{"id":"arxiv/2611.12308","url":"https://arxiv.org/abs/2611.12308","tags":["math.NA"],"score":0,"title":"Unmasking shortcut solutions of coding agents on partial differential equations","domain":"math.NA","source":"arxiv","authors":["Jiale Linghu","Yangshuai Wang"],"reviews":0,"subject":"math.NA","summary":"Coding agents now write simulation programs, but benchmarks usually judge only whether their output matches a reference. In manufactured-solution tasks the data determine the exact solution, so a matching output need not come from a solver. We introduce PDEViBench, a benchmark of 290 tasks in six equation families that traces each passing program to the computation behind its output and reruns unclear programs on changed data. A Codex agent (GPT-5.2) passed 80% of 1,450 runs, yet about one pass in ten returned the exact solution without solving the equation. Counting only solved passes…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-11-09","reactions":{}}