# MixRoute: Rethinking Single-Distribution Training for Generalizable Neural Routing

**Authors:** Hang Yi, Ziwei Huang, Yining Ma, Zhiguang Cao  
**Published:** 9 Nov 2026  
**Links:** [arXiv](https://arxiv.org/abs/2611.11977) · [PDF](https://arxiv.org/pdf/2611.11977) · [This page](https://omen-bit.github.io/ai-paper-daily/paper/2611.11977.html)

## Key numbers

- **179**: we first construct a comprehensive benchmark spanning three representative routing problems and 179 fine-grained datasets

## Abstract

Neural Combinatorial Optimization (NCO) is a promising paradigm for solving routing problems, yet its generalization across diverse data distributions remains a critical bottleneck. Existing methods typically tackle cross-distribution generalization through complex multi-distribution training or intricate meta-learning schemes, implicitly presuming that single-distribution training yields a weak zero-shot baseline. In this paper, we demonstrate that the potential of single-distribution training has been substantially underestimated and can be unlocked by a simple architectural inductive bias: Mix Normalization. To approach this, we first construct a comprehensive benchmark spanning three representative routing problems and 179 fine-grained datasets, capturing diverse shifts in node coordinates, customer demands, capacity and time windows. Building on this benchmark, we propose MixRoute, a neural framework that adaptively combines normalization statistics at different granularities to mitigate varying distribution shifts. Trained exclusively on uniform distributions, MixRoute generalizes effectively to a wide range of unseen distributions without any distribution-specific adaptation. Extensive experiments demonstrate that MixRoute consistently achieves state-of-the-art zero-shot generalization. Our work revisits normalization as a parsimonious alternative for zero-shot generalization against complex training-adaptation schemes.

---

Notes written by AI from the paper, numbers copied from the abstract. Check the paper before you rely on them. From AI Papers Daily.