Evolutionary, Neural, or LLM-Driven Heuristic Generation? A Unified Ant Colony Optimization Benchmark for Nature-Inspired Routing Heuristics on the TSP and CVRP
Haoyuan Wu, You Wu · 2026
Read Evolutionary, Neural, or LLM-Driven Heuristic Generation? A Unified Ant Colony Optimization Benchmark for Nature-Inspired Routing Heuristics on the TSP and CVRP by Haoyuan Wu, You Wu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13406755.txt.
Description
Haoyuan Wu and You Wu benchmark ways of inventing routing rules inside an ant colony solver, in a 2026 computing study. Artificial ants find routes by reinforcing virtual scent trails, a trick borrowed from real colonies. Human rules, evolved programs, a small neural network and a language-model-driven method were compared.
All four generators plugged into the same solver and faced held-out travelling salesman and vehicle routing problems. The evolved program found the shortest routes at every scale tested. The language-model proxy and strong human rules formed a second tier, while the small neural proxy faded as problems grew. The authors caution that constrained stand-ins cannot rank the full paradigms fairly.
Catalog record
| CREATOR | Haoyuan Wu, You Wu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44340 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |