Benchmark RadarAI BENCHMARK PROFILE
CITYREP
CityRep is a benchmark for urban representation learning across 8 cities and 8 tasks with spatially structured splits to mitigate leakage.
- Released
- 2026-05-25
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
It addresses inconsistencies in urban representation evaluation and supports generalization-aware model selection.
Motivation
Urban representation learning encodes complex urban environments into general-purpose embeddings for diverse downstream tasks and emerging urban foundation models.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.