Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require synthesizing clues across large external knowledge sources. As a result, subtle errors can propagate through reasoning traces, potentially never to be detected. Prior work has proposed process reward models (PRMs), including retrieval-augmented variants, but these methods operate post hoc, scoring completed trajectories, which prevents their integration into dynamic inference procedures. Here, we introduce Process Reward Agents (PRA), a test-time method for providing domain-grounded, online, step-wise rewards to a frozen policy. In contrast to prior retrieval-augmented PRMs, PRA enables search-based decoding to rank and prune candidate trajectories at every generation step. Experiments on multiple medical reasoning benchmarks demonstrate that PRA consistently outperforms strong baselines, achieving 80.8% accuracy on MedQA with Qwen3-4B, a new state of the art at the 4B scale. Importantly, PRA generalizes to unseen frozen policy models ranging from 0.5B to 8B parameters, improving their accuracy by up to 25.7% without any policy model updates. More broadly, PRA suggests a paradigm in which frozen reasoners are decoupled from domain-specific reward modules, allowing the deployment of new backbones in complex domains without retraining.

翻译：知识密集型领域的推理仍然具有挑战性，因为中间步骤通常无法进行局部验证：与数学或代码不同，评估步骤的正确性可能需要综合来自大量外部知识源的线索。因此，细微的错误可能会在推理轨迹中传播，并且可能永远无法被检测到。先前的工作提出了过程奖励模型（PRMs），包括检索增强的变体，但这些方法是在事后进行操作，对完成的轨迹进行评分，这阻碍了它们集成到动态推理过程中。在此，我们引入了过程奖励智能体（PRA），这是一种在测试时为冻结策略提供基于领域、在线、逐步奖励的方法。与先前的检索增强型PRMs相比，PRA支持基于搜索的解码，以便在每个生成步骤中对候选轨迹进行排序和剪枝。在多个医学推理基准上的实验表明，PRA持续优于强基线方法，在Qwen3-4B上于MedQA基准上达到80.8%的准确率，树立了4B规模下的新最优水平。重要的是，PRA可泛化到未见的冻结策略模型（参数规模从0.5B到8B），在无需任何策略模型更新的情况下，将其准确率提升高达25.7%。更广泛地说，PRA提出了一种范式，其中冻结的推理器与特定领域的奖励模块解耦，从而无需重新训练即可在复杂领域部署新的骨干模型。