JaxARC: A High-Performance JAX-based Environment for Abstraction and Reasoning Research
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914277726617600 |
|---|---|
| author | Aadam Verma, Monu Abdel-Mottaleb, Mohamed |
| author_facet | Aadam Verma, Monu Abdel-Mottaleb, Mohamed |
| contents | The Abstraction and Reasoning Corpus (ARC) tests AI systems' ability to perform human-like inductive reasoning from a few demonstration pairs. Existing Gymnasium-based RL environments severely limit experimental scale due to computational bottlenecks. We present JaxARC, an open-source, high-performance RL environment for ARC implemented in JAX. Its functional, stateless architecture enables massive parallelism, achieving 38-5,439x speedup over Gymnasium at matched batch sizes, with peak throughput of 790M steps/second. JaxARC supports multiple ARC datasets, flexible action spaces, composable wrappers, and configuration-driven reproducibility, enabling large-scale RL research previously computationally infeasible. JaxARC is available at https://github.com/aadimator/JaxARC. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_17564 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | JaxARC: A High-Performance JAX-based Environment for Abstraction and Reasoning Research Aadam Verma, Monu Abdel-Mottaleb, Mohamed Artificial Intelligence Machine Learning The Abstraction and Reasoning Corpus (ARC) tests AI systems' ability to perform human-like inductive reasoning from a few demonstration pairs. Existing Gymnasium-based RL environments severely limit experimental scale due to computational bottlenecks. We present JaxARC, an open-source, high-performance RL environment for ARC implemented in JAX. Its functional, stateless architecture enables massive parallelism, achieving 38-5,439x speedup over Gymnasium at matched batch sizes, with peak throughput of 790M steps/second. JaxARC supports multiple ARC datasets, flexible action spaces, composable wrappers, and configuration-driven reproducibility, enabling large-scale RL research previously computationally infeasible. JaxARC is available at https://github.com/aadimator/JaxARC. |
| title | JaxARC: A High-Performance JAX-based Environment for Abstraction and Reasoning Research |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2601.17564 |