扩展训练前,、训练后, 和测试时计算已成为提高法学硕士能力的核心范例。在这项工作, 中，我们确定验证, 确定解决方案, 正确性的能力作为新的缩放轴。为了解锁这一点并证明其有效性,，我们引入了LLM-as-a-Verifier,，这是一个通用验证框架，可以为代理任务提供细粒度的反馈，而无需额外的培训。与标准 LM 法官提示 LLM 为候选解决方案生成离散分数不同，2C LLM 作为验证者计算评分标记 logits 分布的期望，以生成连续分数。这种概率公式使验证能够沿着多个维度: (1) 评分粒度, (2) 重复评估, 和(3) 标准分解进行扩展。特别是,，我们表明缩放评分粒度可以更好地分离正负解决方案,，从而实现更校准的比较。此外，, 缩放重复评估和标准分解始终可以通过减少方差和复杂性来额外提高验证准确性。我们进一步引入了一种经济高效的排名算法，用于使用验证者的 连续分数在候选者中选择最佳解决方案。 LLM-as-a-Verifier 在 Terminal-Bench V2 (86.5%), SWE-Bench Verified (78.2%), RoboRewardBench (87.4%), 和 MedAgentBench (73.3%) 上实现了最先进的性能。除了验证,之外，LLM作为验证者的细粒度信号还可以充当估计任务进度的代理。我们为 Claude Code, 构建了一个扩展，使开发人员能够监控和改进他们自己的代理系统。最后,我们展示了LLM作为验证者可以为RL,提供密集的反馈，提高SAC和GRPO在机器人和数学推理基准上的样本效率。
Scaling pre-training, post-training, and test-time compute have become the central paradigms for improving the capabilities of LLMs. In this work, we identify verification, the ability to determine the correctness of a solution, as a new scaling axis. To unlock this and demonstrate its effectiveness, we introduce LLM-as-a-Verifier, a general-purpose verification framework that provides fine-grained feedback for agentic tasks without requiring additional training. Unlike standard LM judges that prompt LLMs to produce discrete scores for candidate solutions, LLM-as-a-Verifier computes the expectation over the distribution of scoring token logits to generate continuous scores. This probabilistic formulation enables verification to scale along multiple dimensions: (1) score granularity, (2) repeated evaluation, and (3) criteria decomposition. In particular, we show that scaling the scoring granularity leads to better separation between positive and negative solutions, resulting in more calibrated comparisons. Moreover, scaling repeated evaluation and criteria decomposition consistently lead to additional gains in verification accuracy through variance and complexity reduction. We further introduce a cost-efficient ranking algorithm for selecting the best solution among candidates using the verifier的 continuous scores. LLM-as-a-Verifier achieves state-of-the-art performance on Terminal-Bench V2 (86.5%), SWE-Bench Verified (78.2%), RoboRewardBench (87.4%), and MedAgentBench (73.3%). Beyond verification, the fine-grained signals from LLM-as-a-Verifier can also serve as a proxy for estimating task progress. We build an extension for Claude Code, enabling developers to monitor and improve their own agentic systems. Finally, we show that LLM-as-a-Verifier can provide dense feedback for RL, improving the sample efficiency of SAC and GRPO on robotics and mathematical reasoning benchmarks.
科目: 人工智能 (cs.AI); 计算和语言 (cs.CL); 机器学习 (cs.LG); 多代理系统 (cs.MA); 机器人 (cs.RO)
Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Robotics (cs.RO)