GeoAgent：透過強化地理特徵學習實現無所不在的地理定位

摘要

本文提出GeoAgent模型，該模型能夠以貼近人類思維的方式進行推理，並得出細粒度地址結論。儘管先前基於強化學習的方法在效能與可解釋性方面取得突破，但其依賴人工智慧生成的思維鏈數據與訓練策略仍存在隱憂，這些方法與地理特性存在衝突。為解決這些問題，我們首先引入GeoSeek——一個由地理專家和專業玩家共同標註思維鏈數據的新型地理定位數據集。我們進一步深入探究地理任務的內在特性，提出通過一致性智能體評估的地理相似度獎勵與一致性獎勵機制來輔助訓練。這促使模型能從地理視角向正確答案收斂，同時確保其推理過程的完整性和一致性。實驗結果表明，GeoAgent在多個粒度層面上均超越現有方法及一系列通用視覺語言大模型，且生成的推理過程與人類思維高度契合。

English

This paper presents GeoAgent, a model capable of reasoning closely with humans and deriving fine-grained address conclusions. Previous RL-based methods have achieved breakthroughs in performance and interpretability but still remain concerns because of their reliance on AI-generated chain-of-thought (CoT) data and training strategies, which conflict with geographic characteristics. To address these issues, we first introduce GeoSeek, a new geolocation dataset comprising CoT data annotated by geographic experts and professional players. We further thoroughly explore the inherent characteristics of geographic tasks and propose a geo-similarity reward and a consistency reward assessed by a consistency agent to assist training. This encourages the model to converge towards correct answers from a geographic perspective while ensuring the integrity and consistency of its reasoning process. Experimental results show that GeoAgent outperforms existing methods and a series of general VLLMs across multiple grains, while generating reasoning that closely aligns with humans.

GeoAgent：透過強化地理特徵學習實現無所不在的地理定位

GeoAgent: Learning to Geolocate Everywhere with Reinforced Geographic Characteristics

摘要

Support