GTR-Bench: Evaluating Geo-Temporal Reasoning in Vision-Language Models
This addresses a gap in assessing VLMs' spatial-temporal intelligence for applications like traffic management and emergency response, though it is incremental as it builds on existing benchmark efforts.
The paper tackles the lack of benchmarks for evaluating geo-temporal reasoning in vision-language models (VLMs) by introducing GTR-Bench, a novel challenge that requires reasoning across maps and videos with non-overlapping views, and finds that even top models like Gemini-2.5-Pro achieve only 34.9% accuracy compared to human performance of 78.61%.
Recently spatial-temporal intelligence of Visual-Language Models (VLMs) has attracted much attention due to its importance for Autonomous Driving, Embodied AI and General Artificial Intelligence. Existing spatial-temporal benchmarks mainly focus on egocentric perspective reasoning with images/video context, or geographic perspective reasoning with graphics context (eg. a map), thus fail to assess VLMs' geographic spatial-temporal intelligence with both images/video and graphics context, which is important for areas like traffic management and emergency response. To address the gaps, we introduce Geo-Temporal Reasoning benchmark (GTR-Bench), a novel challenge for geographic temporal reasoning of moving targets in a large-scale camera network. GTR-Bench is more challenging as it requires multiple perspective switches between maps and videos, joint reasoning across multiple videos with non-overlapping fields of view, and inference over spatial-temporal regions that are unobserved by any video context. Evaluations of more than 10 popular VLMs on GTR-Bench demonstrate that even the best proprietary model, Gemini-2.5-Pro (34.9%), significantly lags behind human performance (78.61%) on geo-temporal reasoning. Moreover, our comprehensive analysis on GTR-Bench reveals three primary deficiencies of current models for geo-temporal reasoning. (1) VLMs' reasoning is impaired by an imbalanced utilization of spatial-temporal context. (2) VLMs are weak in temporal forecasting, which leads to worse performance on temporal-emphasized tasks than on spatial-emphasized tasks. (3) VLMs lack the proficiency to comprehend or align the map data with multi-view video inputs. We believe GTR-Bench offers valuable insights and opens up new opportunities for research and applications in spatial-temporal intelligence. Benchmark and code will be released at https://github.com/X-Luffy/GTR-Bench.