TimeSpot: Benchmarking Geo-Temporal Understanding in Vision–Language Models in Real-World Settings

Azmine Toushik Wasi, Shahriyar Zaman Ridoy, Koushik Ahamed Tonmoy, Kinga Tshering, S. M. Muhtasimul Hasan, Wahid Faisal, Tasnim Mohiuddin, Md Rizwan Parvez
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:133247-133304, 2026.

Abstract

Geo-temporal understanding, the ability to infer location, time, and contextual properties from visual input alone, underpins applications such as disaster management, traffic planning, embodied navigation, world modeling, and geography education. Although recent vision–language models (VLMs) have advanced image geo-localization using cues like landmarks and road signs, their ability to reason about temporal signals and physically grounded spatial cues remains limited. To address this gap, we introduce TimeSpot, a benchmark for evaluating real-world geo-temporal reasoning in VLMs. TimeSpot comprises 1,455 ground-level images from 80 countries and requires structured prediction of temporal attributes (season, month, time of day, daylight phase) and geographic attributes (continent, country, climate zone, environment type, latitude–longitude) directly from visual evidence. It also includes spatial–temporal reasoning tasks that test physical plausibility under real-world uncertainty. Evaluations of state-of-the-art open- and closed-source VLMs show low performance, particularly for temporal inference. While supervised fine-tuning yields improvements, results remain insufficient, highlighting the need for new methods to achieve robust, physically grounded geo-temporal understanding. TimeSpot is available at: https://TimeSpot-GT.github.io.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-wasi26a, title = {{T}ime{S}pot: Benchmarking Geo-Temporal Understanding in Vision–Language Models in Real-World Settings}, author = {Wasi, Azmine Toushik and Ridoy, Shahriyar Zaman and Tonmoy, Koushik Ahamed and Tshering, Kinga and Hasan, S. M. Muhtasimul and Faisal, Wahid and Mohiuddin, Tasnim and Parvez, Md Rizwan}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {133247--133304}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/wasi26a/wasi26a.pdf}, url = {https://proceedings.mlr.press/v306/wasi26a.html}, abstract = {Geo-temporal understanding, the ability to infer location, time, and contextual properties from visual input alone, underpins applications such as disaster management, traffic planning, embodied navigation, world modeling, and geography education. Although recent vision–language models (VLMs) have advanced image geo-localization using cues like landmarks and road signs, their ability to reason about temporal signals and physically grounded spatial cues remains limited. To address this gap, we introduce TimeSpot, a benchmark for evaluating real-world geo-temporal reasoning in VLMs. TimeSpot comprises 1,455 ground-level images from 80 countries and requires structured prediction of temporal attributes (season, month, time of day, daylight phase) and geographic attributes (continent, country, climate zone, environment type, latitude–longitude) directly from visual evidence. It also includes spatial–temporal reasoning tasks that test physical plausibility under real-world uncertainty. Evaluations of state-of-the-art open- and closed-source VLMs show low performance, particularly for temporal inference. While supervised fine-tuning yields improvements, results remain insufficient, highlighting the need for new methods to achieve robust, physically grounded geo-temporal understanding. TimeSpot is available at: https://TimeSpot-GT.github.io.} }
Endnote
%0 Conference Paper %T TimeSpot: Benchmarking Geo-Temporal Understanding in Vision–Language Models in Real-World Settings %A Azmine Toushik Wasi %A Shahriyar Zaman Ridoy %A Koushik Ahamed Tonmoy %A Kinga Tshering %A S. M. Muhtasimul Hasan %A Wahid Faisal %A Tasnim Mohiuddin %A Md Rizwan Parvez %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-wasi26a %I PMLR %P 133247--133304 %U https://proceedings.mlr.press/v306/wasi26a.html %V 306 %X Geo-temporal understanding, the ability to infer location, time, and contextual properties from visual input alone, underpins applications such as disaster management, traffic planning, embodied navigation, world modeling, and geography education. Although recent vision–language models (VLMs) have advanced image geo-localization using cues like landmarks and road signs, their ability to reason about temporal signals and physically grounded spatial cues remains limited. To address this gap, we introduce TimeSpot, a benchmark for evaluating real-world geo-temporal reasoning in VLMs. TimeSpot comprises 1,455 ground-level images from 80 countries and requires structured prediction of temporal attributes (season, month, time of day, daylight phase) and geographic attributes (continent, country, climate zone, environment type, latitude–longitude) directly from visual evidence. It also includes spatial–temporal reasoning tasks that test physical plausibility under real-world uncertainty. Evaluations of state-of-the-art open- and closed-source VLMs show low performance, particularly for temporal inference. While supervised fine-tuning yields improvements, results remain insufficient, highlighting the need for new methods to achieve robust, physically grounded geo-temporal understanding. TimeSpot is available at: https://TimeSpot-GT.github.io.
APA
Wasi, A.T., Ridoy, S.Z., Tonmoy, K.A., Tshering, K., Hasan, S.M.M., Faisal, W., Mohiuddin, T. & Parvez, M.R.. (2026). TimeSpot: Benchmarking Geo-Temporal Understanding in Vision–Language Models in Real-World Settings. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:133247-133304 Available from https://proceedings.mlr.press/v306/wasi26a.html.

Related Material