TurnBack: A Geospatial Route Cognition Benchmark for Large Language Models through Reverse Route
This addresses the underexplored geospatial cognition capabilities of LLMs for researchers in AI and natural language processing, though it is incremental as it builds on prior work with new datasets and metrics.
The authors tackled the problem of evaluating geospatial route cognition in large language models (LLMs) by creating a large-scale benchmark with 36,000 routes from 12 metropolises, revealing that LLMs struggle with route reversal, often failing to return to the starting point or match optimal routes, with low robustness and high confidence in incorrect answers.
Humans can interpret geospatial information through natural language, while the geospatial cognition capabilities of Large Language Models (LLMs) remain underexplored. Prior research in this domain has been constrained by non-quantifiable metrics, limited evaluation datasets and unclear research hierarchies. Therefore, we propose a large-scale benchmark and conduct a comprehensive evaluation of the geospatial route cognition of LLMs. We create a large-scale evaluation dataset comprised of 36000 routes from 12 metropolises worldwide. Then, we introduce PathBuilder, a novel tool for converting natural language instructions into navigation routes, and vice versa, bridging the gap between geospatial information and natural language. Finally, we propose a new evaluation framework and metrics to rigorously assess 11 state-of-the-art (SOTA) LLMs on the task of route reversal. The benchmark reveals that LLMs exhibit limitation to reverse routes: most reverse routes neither return to the starting point nor are similar to the optimal route. Additionally, LLMs face challenges such as low robustness in route generation and high confidence for their incorrect answers. Code\ \&\ Data available here: \href{https://github.com/bghjmn32/EMNLP2025_Turnback}{TurnBack.}