Exploring Geographic Relative Space in Large Language Models through Activation Patching
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866917494536536064 |
|---|---|
| author | De Sabbata, Stef Baiju, Rahul Mizzaro, Stefano Roitero, Kevin |
| author_facet | De Sabbata, Stef Baiju, Rahul Mizzaro, Stefano Roitero, Kevin |
| contents | The increased use of Large Language Models (LLMs) in geography raises substantial questions about the safety of integrating these tools across a wide range of processes and analyses, given our very limited understanding of their inner workings. In this extended abstract, we examine how LLMs process relative geographic space using activation patching, an emerging tool for mechanistic interpretability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_14535 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Exploring Geographic Relative Space in Large Language Models through Activation Patching De Sabbata, Stef Baiju, Rahul Mizzaro, Stefano Roitero, Kevin Machine Learning The increased use of Large Language Models (LLMs) in geography raises substantial questions about the safety of integrating these tools across a wide range of processes and analyses, given our very limited understanding of their inner workings. In this extended abstract, we examine how LLMs process relative geographic space using activation patching, an emerging tool for mechanistic interpretability. |
| title | Exploring Geographic Relative Space in Large Language Models through Activation Patching |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2605.14535 |