Multilingual Reasoning Gym: Multilingual Scaling of Procedural Reasoning Environments
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910049303003136 |
|---|---|
| author | Dobler, Konstantin Lehnerer, Simon Scozzafava, Federico Janke, Jonathan Ali, Mohamed |
| author_facet | Dobler, Konstantin Lehnerer, Simon Scozzafava, Federico Janke, Jonathan Ali, Mohamed |
| contents | We present the Multilingual Reasoning Gym, an extension of Reasoning Gym (Stojanovski et al., 2025), that procedurally generates verifiable reasoning problems across 14 languages. We translate templates for 94 tasks with native-speaker validation in 10 languages and targeted code or template adaptations to ensure linguistic naturalness. The Multilingual Reasoning Gym preserves the core benefits of the procedural generation approach used in the original Reasoning Gym, such as virtually unlimited problem instance generation and adjustable difficulty, and remains directly usable for Reinforcement Learning from Verifiable Rewards and evaluation settings. Problems in the Multilingual Reasoning Gym are parallel across languages, enabling crosslingually parallel data generation at massive scale due to the procedural nature of the environments. We release our implementation to support research into multilingual reasoning models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_10793 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Multilingual Reasoning Gym: Multilingual Scaling of Procedural Reasoning Environments Dobler, Konstantin Lehnerer, Simon Scozzafava, Federico Janke, Jonathan Ali, Mohamed Computation and Language We present the Multilingual Reasoning Gym, an extension of Reasoning Gym (Stojanovski et al., 2025), that procedurally generates verifiable reasoning problems across 14 languages. We translate templates for 94 tasks with native-speaker validation in 10 languages and targeted code or template adaptations to ensure linguistic naturalness. The Multilingual Reasoning Gym preserves the core benefits of the procedural generation approach used in the original Reasoning Gym, such as virtually unlimited problem instance generation and adjustable difficulty, and remains directly usable for Reinforcement Learning from Verifiable Rewards and evaluation settings. Problems in the Multilingual Reasoning Gym are parallel across languages, enabling crosslingually parallel data generation at massive scale due to the procedural nature of the environments. We release our implementation to support research into multilingual reasoning models. |
| title | Multilingual Reasoning Gym: Multilingual Scaling of Procedural Reasoning Environments |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2603.10793 |