Comment on paper: Position: Rethinking Post-Hoc Search-Based Neural Approaches for Solving Large-Scale Traveling Salesman Problems
This is an incremental critique that addresses methodological flaws in a prior study on solving large-scale traveling salesman problems.
The authors identified two major issues in the SoftDist paper, including inconsistent hardware environments and time measurements, which led to flawed conclusions; when corrected, the primary claim of SoftDist is no longer supported.
We identify two major issues in the SoftDist paper (Xia et al.): (1) the failure to run all steps of different baselines on the same hardware environment, and (2) the use of inconsistent time measurements when comparing to other baselines. These issues lead to flawed conclusions. When all steps are executed in the same hardware environment, the primary claim made in SoftDist is no longer supported.