SpotAgent: Grounding Visual Geo-localization in Large Vision-Language Models through Agentic Reasoning

TL;DR

SpotAgent introduces an agentic reasoning framework for visual geo-localization that combines large vision-language models with external tools and a multi-stage training pipeline, achieving state-of-the-art accuracy and verifiability.

Contribution

The paper presents a novel agentic reasoning approach with a multi-stage training pipeline and dynamic filtering, improving geo-localization accuracy and verifiability over prior methods.

Findings

Achieves state-of-the-art performance on geo-localization benchmarks.

Effectively reduces hallucinations and improves result verifiability.

Enhances reasoning with external tools and a specialized training strategy.

Abstract

Large Vision-Language Models (LVLMs) have demonstrated strong reasoning capabilities in geo-localization, yet they often struggle in real-world scenarios where visual cues are sparse, long-tailed, and highly ambiguous. Previous approaches, bound by internal knowledge, often fail to provide verifiable results, yielding confident but ungrounded predictions when faced with confounded evidence. To address these challenges, we propose SpotAgent, a framework that formalizes geo-localization into an agentic reasoning process that leverages expert-level reasoning to synergize visual interpretation with tool-assisted verification. SpotAgent actively explores and verifies visual cues by leveraging external tools (e.g., web search, maps) through a ReAct diagram. We introduce a 3-stage post-training pipeline starting with a Supervised Fine-Tuning (SFT) stage for basic alignment, followed by an Agentic Cold Start phase utilizing high-quality trajectories synthesized via a Multi-Agent framework, aiming to instill tool-calling expertise. Subsequently, the model's reasoning capabilities are refined through Reinforcement Learning. We propose a Spatially-Aware Dynamic Filtering strategy to enhance the efficiency of the RL stage by prioritizing learnable samples based on spatial difficulty. Extensive experiments on standard benchmarks demonstrate that SpotAgent achieves state-of-the-art performance, effectively mitigating hallucinations while delivering precise and verifiable geo-localization.