Nudging Beyond the Comfort Zone: Efficient Strategy-Guided Exploration for RLVR

Hugging Face Daily Papers Papers

Summary

NudgeRL is a framework that enhances reinforcement learning with verifiable rewards (RLVR) by introducing structured exploration and strategy nudging, achieving better reasoning performance in large language models more efficiently than brute-force scaling methods.

Reinforcement learning with verifiable rewards (RLVR) has emerged as a scalable paradigm for improving the reasoning capabilities of large language models. However, its effectiveness is fundamentally limited by exploration: the policy can only improve on trajectories it has already sampled. While increasing the number of rollouts alleviates this issue, such brute-force scaling is computationally expensive, and existing approaches that modify the optimization objective provide limited control over what is explored. In this work, we propose NudgeRL, a framework for structured and diversity-driven exploration in RLVR. Our approach introduces Strategy Nudging, which conditions each rollout on lightweight, strategy-level contexts to induce diverse reasoning trajectories without relying on expensive oracle supervision. To effectively learn from such structured exploration, we further propose a unified objective, which decomposes the reward signal into inter- and intra-context components and incorporates a distillation objective to transfer discovered behaviors back to the base policy. Empirically, NudgeRL outperforms standard GRPO with up to 8 times larger rollout budgets, while outperforming oracle-guided RL baseline on average across five challenging math benchmarks. These results demonstrate that structured, context-driven exploration can serve as an efficient and scalable alternative to both brute-force rollout scaling and feasibility-oriented methods based on privileged information. Our code is available at https://github.com/tally0818/NudgeRL.
Original Article
View Cached Full Text

Cached at: 05/18/26, 06:24 AM

Paper page - Nudging Beyond the Comfort Zone: Efficient Strategy-Guided Exploration for RLVR

Source: https://huggingface.co/papers/2605.15726

Abstract

NudgeRL framework enhances reinforcement learning with verifiable rewards through structured exploration and strategy nudging to improve reasoning capabilities in large language models.

Reinforcement learning with verifiable rewards(RLVR) has emerged as a scalable paradigm for improving the reasoning capabilities of large language models. However, its effectiveness is fundamentally limited byexploration: the policy can only improve on trajectories it has already sampled. While increasing the number ofrolloutsalleviates this issue, such brute-force scaling is computationally expensive, and existing approaches that modify the optimization objective provide limited control over what is explored. In this work, we propose NudgeRL, a framework for structured and diversity-drivenexplorationin RLVR. Our approach introduces Strategy Nudging, which conditions each rollout on lightweight,strategy-level contextsto induce diverse reasoning trajectories without relying on expensiveoracle supervision. To effectively learn from such structuredexploration, we further propose a unified objective, which decomposes the reward signal into inter- and intra-context components and incorporates adistillation objectiveto transfer discovered behaviors back to the base policy. Empirically, NudgeRL outperforms standardGRPOwith up to 8 times larger rollout budgets, while outperforming oracle-guided RL baseline on average across five challenging math benchmarks. These results demonstrate that structured, context-drivenexplorationcan serve as an efficient and scalable alternative to both brute-force rollout scaling and feasibility-oriented methods based on privileged information. Our code is available at https://github.com/tally0818/NudgeRL.

View arXiv pageView PDFGitHub2Add to collection

Get this paper in your agent:

hf papers read 2605\.15726

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.15726 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.15726 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.15726 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Locked at the Entrance, Open Inside: Where RLVR Narrows the Solution Space

arXiv cs.LG

This paper investigates how reinforcement learning with verifiable rewards (RLVR) narrows the solution space in LLM reasoning by analyzing where diversity is lost, finding it concentrated at the 'entrance' of trajectories. It demonstrates interventions to recover breadth without compromising accuracy.

ExpRL: Exploratory RL for LLM Mid-Training

Hugging Face Daily Papers

ExpRL is a new RL-based mid-training method that uses human-written reference solutions as dense reward scaffolds (never shown to the policy) to improve LLM reasoning, achieving significant gains on hard math benchmarks like AIME-2026.