Search papers, labs, and topics across Lattice.
The authors introduce the Multilingual Reasoning Gym, an extension of the Reasoning Gym, which procedurally generates verifiable reasoning problems across 14 languages. They translated templates for 94 tasks, validated by native speakers, and adapted code/templates to ensure linguistic naturalness. This new gym maintains the benefits of procedural generation, like unlimited problem instances and adjustable difficulty, while enabling crosslingually parallel data generation for training and evaluation.
Unlock massive multilingual reasoning data: the Multilingual Reasoning Gym enables parallel data generation across 14 languages, opening doors for training and evaluating multilingual reasoning models at scale.
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.
