Are Video Generation Models Geographically Fair? An Attraction-Centric Evaluation of Global Visual Knowledge

2026-01-26

AI Summary

This study investigates whether text-to-video generation models exhibit geographic fairness—i.e., the ability to equitably represent visual knowledge across diverse global regions. To this end, the authors propose the Geo-Attraction Landmark Probing (GAP) evaluation framework and introduce GEOATTRACTION-500, a benchmark comprising 500 geographically distributed landmarks, enabling the first quantitative assessment of geographic fairness in such models. The framework integrates multidimensional metrics, including global structural alignment, fine-grained keypoint matching, and vision-language model scoring, and demonstrates strong consistency with human evaluations. Experiments on Sora 2 reveal a relatively uniform distribution of geographic visual knowledge, with minimal bias across regions, levels of economic development, and cultural groups, and limited influence from landmark popularity.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Recent advances in text-to-video generation have produced visually compelling results, yet it remains unclear whether these models encode geographically equitable visual knowledge. In this work, we investigate the geo-equity and geographically grounded visual knowledge of text-to-video models through an attraction-centric evaluation. We introduce Geo-Attraction Landmark Probing (GAP), a systematic framework for assessing how faithfully models synthesize tourist attractions from diverse regions, and construct GEOATTRACTION-500, a benchmark of 500 globally distributed attractions spanning varied regions and popularity levels. GAP integrates complementary metrics that disentangle overall video quality from attraction-specific knowledge, including global structural alignment, fine-grained keypoint-based alignment, and vision-language model judgments, all validated against human evaluation. Applying GAP to the state-of-the-art text-to-video model Sora 2, we find that, contrary to common assumptions of strong geographic bias, the model exhibits a relatively uniform level of geographically grounded visual knowledge across regions, development levels, and cultural groupings, with only weak dependence on attraction popularity. These results suggest that current text-to-video models express global visual knowledge more evenly than expected, highlighting both their promise for globally deployed applications and the need for continued evaluation as such systems evolve.

Problem

• geographic fairness
• video generation
• visual knowledge
• global representation
• text-to-video models

Innovation

• geo-equity
• text-to-video generation
• landmark evaluation
• visual knowledge fairness
• GAP framework