PubSwap: Public-Data Off-Policy Coordination for Federated RLVR

TL;DR

This paper introduces PubSwap, a federated RLVR framework that uses public data and low-rank adaptation to improve communication efficiency and coordination across decentralized organizations.

Contribution

It proposes a novel combination of LoRA-based local adaptation and public-data off-policy steps for scalable, privacy-preserving federated reinforcement learning from verifiable rewards.

Findings

Consistently improves performance on mathematical and medical reasoning benchmarks.

Enhances communication efficiency and cross-client coordination in federated RLVR.

Demonstrates effectiveness of combining low-rank adaptation with public-data exchange.

Abstract

Reasoning post-training with reinforcement learning from verifiable rewards (RLVR) is typically studied in centralized settings, yet many realistic applications involve decentralized private data distributed across organizations. Federated training is a natural solution, but scaling RLVR in this regime is challenging: full-model synchronization is expensive, and performing many local steps can cause severe client drift under heterogeneous data. We propose a federated RLVR framework that combines LoRA-based local adaptation with public-data-based off-policy steps to improve both communication efficiency and cross-client coordination. In particular, a small shared public dataset is used to periodically exchange and reuse response-level training signals across organizations, providing a lightweight anchor toward a more globally aligned objective without exposing private data. Our method selectively replaces locally incorrect responses with globally correct ones during public-data steps, thereby keeping training closer to the local policy while still benefiting from cross-client coordination. Across mathematical and medical reasoning benchmarks and models, our method consistently improves over standard baselines. Our results highlight a simple and effective recipe for federated reasoning post-training: combining low-rank communication with limited public-data coordination.