ARC Is a Vision Problem!
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908662897836032 |
|---|---|
| author | Hu, Keya Cy, Ali Qiu, Linlu Ding, Xiaoman Delores Wang, Runqian Zhu, Yeyin Eva Andreas, Jacob He, Kaiming |
| author_facet | Hu, Keya Cy, Ali Qiu, Linlu Ding, Xiaoman Delores Wang, Runqian Zhu, Yeyin Eva Andreas, Jacob He, Kaiming |
| contents | The Abstraction and Reasoning Corpus (ARC) is designed to promote research on abstract reasoning, a fundamental aspect of human intelligence. Common approaches to ARC treat it as a language-oriented problem, addressed by large language models (LLMs) or recurrent reasoning models. However, although the puzzle-like tasks in ARC are inherently visual, existing research has rarely approached the problem from a vision-centric perspective. In this work, we formulate ARC within a vision paradigm, framing it as an image-to-image translation problem. To incorporate visual priors, we represent the inputs on a "canvas" that can be processed like natural images. It is then natural for us to apply standard vision architectures, such as a vanilla Vision Transformer (ViT), to perform image-to-image mapping. Our model is trained from scratch solely on ARC data and generalizes to unseen tasks through test-time training. Our framework, termed Vision ARC (VARC), achieves 60.4% accuracy on the ARC-1 benchmark, substantially outperforming existing methods that are also trained from scratch. Our results are competitive with those of leading LLMs and close the gap to average human performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_14761 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ARC Is a Vision Problem! Hu, Keya Cy, Ali Qiu, Linlu Ding, Xiaoman Delores Wang, Runqian Zhu, Yeyin Eva Andreas, Jacob He, Kaiming Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning The Abstraction and Reasoning Corpus (ARC) is designed to promote research on abstract reasoning, a fundamental aspect of human intelligence. Common approaches to ARC treat it as a language-oriented problem, addressed by large language models (LLMs) or recurrent reasoning models. However, although the puzzle-like tasks in ARC are inherently visual, existing research has rarely approached the problem from a vision-centric perspective. In this work, we formulate ARC within a vision paradigm, framing it as an image-to-image translation problem. To incorporate visual priors, we represent the inputs on a "canvas" that can be processed like natural images. It is then natural for us to apply standard vision architectures, such as a vanilla Vision Transformer (ViT), to perform image-to-image mapping. Our model is trained from scratch solely on ARC data and generalizes to unseen tasks through test-time training. Our framework, termed Vision ARC (VARC), achieves 60.4% accuracy on the ARC-1 benchmark, substantially outperforming existing methods that are also trained from scratch. Our results are competitive with those of leading LLMs and close the gap to average human performance. |
| title | ARC Is a Vision Problem! |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2511.14761 |