RLVE: Scaling Up Reinforcement Learning for Language Models with Adaptive Verifiable Environments

TL;DR

RLVE introduces adaptive, verifiable environments for reinforcement learning, enabling scalable training of language models that significantly improve reasoning capabilities over static data methods.

Contribution

The paper presents RLVE, a novel approach with a large suite of environments that adapt difficulty during training, enhancing reasoning skills in language models.

Findings

Environment scaling improves reasoning generalization.

RLVE achieves 3.37% gain on benchmarks, outperforming traditional RL.

Code is publicly released for reproducibility.

Abstract

We introduce Reinforcement Learning (RL) with Adaptive Verifiable Environments (RLVE), an approach using verifiable environments that procedurally generate problems and provide algorithmically verifiable rewards, to scale up RL for language models (LMs). RLVE enables each verifiable environment to dynamically adapt its problem difficulty distribution to the policy model's capabilities as training progresses. In contrast, static data distributions often lead to vanishing learning signals when problems are either too easy or too hard for the policy. To implement RLVE, we create RLVE-Gym, a large-scale suite of 400 verifiable environments carefully developed through manual environment engineering. Using RLVE-Gym, we show that environment scaling, i.e., expanding the collection of training environments, consistently improves generalizable reasoning capabilities. RLVE with joint training across all 400 environments in RLVE-Gym yields a 3.37% absolute average improvement across six reasoning benchmarks, starting from one of the strongest 1.5B reasoning LMs. By comparison, continuing this LM's original RL training yields only a 0.49% average absolute gain despite using over 3x more compute. We release our code publicly.