{"id":"arxiv/2608.12893","url":"https://arxiv.org/abs/2608.12893","tags":["math.NA"],"score":0,"title":"A Counterexample to Robust Second-Order Convergence of the Strang Projector-Splitting Integrator","domain":"math.NA","source":"arxiv","authors":["Shiheng Zhang"],"reviews":0,"subject":"math.NA","summary":"The classical Strang projector-splitting integrator is widely observed to converge with order two, whereas the error analysis that remains uniform as the smallest singular value retained in the low-rank approximation tends to zero proves only order one. We show that this gap is intrinsic under the standard assumptions. We construct $3\\times3$ matrix differential equations that are $C^2$ in time and smooth in the matrix variable, with rank-two initial data that satisfy uniform boundedness, Lipschitz, tangency-defect, and regularity bounds. Nevertheless, the exact-subflow Strang method has a…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-08-13","reactions":{}}