Comment on paper: Position: Rethinking Post-Hoc Search-Based Neural Approaches for Solving Large-Scale Traveling Salesman Problems

TL;DR

This paper critically examines the SoftDist approach for large-scale TSP, highlighting experimental inconsistencies and demonstrating that its main claims do not hold under controlled, uniform hardware conditions.

Contribution

It identifies key methodological flaws in prior work, emphasizing the importance of consistent experimental setup for valid comparisons in neural TSP solutions.

Findings

Inconsistent hardware environments affected prior results

Uniform testing invalidates previous main claims

Highlights need for standardized benchmarking

Abstract

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.