GenMRP: A Generative Multi-Route Planning Framework for Efficient and Personalized Real-Time Industrial Navigation

Chengzhang Wang, Chao Chen, Jun Tao, Tengfei Liu, He Bai, Song Wang, Longfei Xu, Kaikui Liu, Xiangxiang Chu

Type: dataset
Venue: AMAP-ML (Amap, Alibaba Group); Southern University of Science and Technology
Year: 2026
Source: huggingface
Access: free
Added: 2026-07-17T20:18:03.733169+00:00
Verified: 2026-07-17T20:18:03.733169+00:00

Summary

GenMRP is a dataset for generative multi-route planning in industrial-scale navigation, released alongside the GenMRP framework from Alibaba's Amap team. It contains ~329K rows of road network features, user historical route sequences, and route planning data covering optimal, alternative, and personalized route planning tasks. The framework uses a skeleton-to-capillary approach with iterative route generation and correctional boosting.

Keywords

route-planning navigation graph transportation personalization amap alibaba generative

Topics

Graph Machine Learning / Route Planning

Research notes

Paper: arxiv 2602.04174. Published by Alibaba's Amap (AMAP-ML) team. No explicit license stated on the dataset page.