@inproceedings{luo-etal-2025-turnback,
title = "{T}urn{B}ack: A Geospatial Route Cognition Benchmark for Large Language Models through Reverse Route",
author = "Luo, Hongyi and
Cheng, Qing and
Matos, Daniel and
Gadi, Hari Krishna and
Zhang, Yanfeng and
Liu, Lu and
Wang, Yongliang and
Zeller, Niclas and
Cremers, Daniel and
Meng, Liqiu",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.emnlp-main.440/",
doi = "10.18653/v1/2025.emnlp-main.440",
pages = "8710--8729",
ISBN = "979-8-89176-332-6",
abstract = "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; unclear research hierarchies further compound these limitations. Therefore, we propose a scalable 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. 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 9 state-of-the-art (SOTA) LLMs, on the task of route reversal. The benchmark reveals that LLMs exhibit limited ability to reverse routes: most of the 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."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="luo-etal-2025-turnback">
<titleInfo>
<title>TurnBack: A Geospatial Route Cognition Benchmark for Large Language Models through Reverse Route</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hongyi</namePart>
<namePart type="family">Luo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qing</namePart>
<namePart type="family">Cheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Matos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hari</namePart>
<namePart type="given">Krishna</namePart>
<namePart type="family">Gadi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanfeng</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yongliang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Niclas</namePart>
<namePart type="family">Zeller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Cremers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liqiu</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-332-6</identifier>
</relatedItem>
<abstract>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; unclear research hierarchies further compound these limitations. Therefore, we propose a scalable 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. 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 9 state-of-the-art (SOTA) LLMs, on the task of route reversal. The benchmark reveals that LLMs exhibit limited ability to reverse routes: most of the 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.</abstract>
<identifier type="citekey">luo-etal-2025-turnback</identifier>
<identifier type="doi">10.18653/v1/2025.emnlp-main.440</identifier>
<location>
<url>https://aclanthology.org/2025.emnlp-main.440/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>8710</start>
<end>8729</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TurnBack: A Geospatial Route Cognition Benchmark for Large Language Models through Reverse Route
%A Luo, Hongyi
%A Cheng, Qing
%A Matos, Daniel
%A Gadi, Hari Krishna
%A Zhang, Yanfeng
%A Liu, Lu
%A Wang, Yongliang
%A Zeller, Niclas
%A Cremers, Daniel
%A Meng, Liqiu
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-332-6
%F luo-etal-2025-turnback
%X 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; unclear research hierarchies further compound these limitations. Therefore, we propose a scalable 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. 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 9 state-of-the-art (SOTA) LLMs, on the task of route reversal. The benchmark reveals that LLMs exhibit limited ability to reverse routes: most of the 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.
%R 10.18653/v1/2025.emnlp-main.440
%U https://aclanthology.org/2025.emnlp-main.440/
%U https://doi.org/10.18653/v1/2025.emnlp-main.440
%P 8710-8729
Markdown (Informal)
[TurnBack: A Geospatial Route Cognition Benchmark for Large Language Models through Reverse Route](https://aclanthology.org/2025.emnlp-main.440/) (Luo et al., EMNLP 2025)
ACL
- Hongyi Luo, Qing Cheng, Daniel Matos, Hari Krishna Gadi, Yanfeng Zhang, Lu Liu, Yongliang Wang, Niclas Zeller, Daniel Cremers, and Liqiu Meng. 2025. TurnBack: A Geospatial Route Cognition Benchmark for Large Language Models through Reverse Route. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 8710–8729, Suzhou, China. Association for Computational Linguistics.