REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards
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_ | 1866911221767208960 |
|---|---|
| author | Stojanovski, Zafir Stanley, Oliver Sharratt, Joe Jones, Richard Adefioye, Abdulhakeem Kaddour, Jean Köpf, Andreas |
| author_facet | Stojanovski, Zafir Stanley, Oliver Sharratt, Joe Jones, Richard Adefioye, Abdulhakeem Kaddour, Jean Köpf, Andreas |
| contents | We introduce Reasoning Gym (RG), a library of reasoning environments for reinforcement learning with verifiable rewards. It provides over 100 data generators and verifiers spanning multiple domains including algebra, arithmetic, computation, cognition, geometry, graph theory, logic, and various common games. Its key innovation is the ability to generate virtually infinite training data with adjustable complexity, unlike most previous reasoning datasets, which are typically fixed. This procedural generation approach allows for continuous evaluation across varying difficulty levels. Our experimental results demonstrate the efficacy of RG in both evaluating and reinforcement learning of reasoning models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_24760 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards Stojanovski, Zafir Stanley, Oliver Sharratt, Joe Jones, Richard Adefioye, Abdulhakeem Kaddour, Jean Köpf, Andreas Machine Learning Artificial Intelligence Computation and Language We introduce Reasoning Gym (RG), a library of reasoning environments for reinforcement learning with verifiable rewards. It provides over 100 data generators and verifiers spanning multiple domains including algebra, arithmetic, computation, cognition, geometry, graph theory, logic, and various common games. Its key innovation is the ability to generate virtually infinite training data with adjustable complexity, unlike most previous reasoning datasets, which are typically fixed. This procedural generation approach allows for continuous evaluation across varying difficulty levels. Our experimental results demonstrate the efficacy of RG in both evaluating and reinforcement learning of reasoning models. |
| title | REASONING GYM: Reasoning Environments for Reinforcement Learning with Verifiable Rewards |
| topic | Machine Learning Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2505.24760 |