推理时的验证扩展：通过测试时的评分指南验证实现自我进化深度研究代理
Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification
AI总结：
本文提出通过测试时评分指南验证实现深度研究代理的自我进化，通过迭代验证策略模型输出，提升验证能力，实验显示在GAIA和XBench-DeepSearch上准确率提升8%-11%。
摘要：
近年来，深度研究代理（DRAs）的进步正在改变自动化知识发现和问题解决。尽管现有研究主要集中在通过训练后增强策略能力，我们提出了一种替代范式：通过迭代验证策略模型输出，由精心设计的评分指南引导，从而实现推理时的验证扩展。这种方法使代理能够通过评估生成答案来自我改进，产生迭代反馈和改进。我们基于自动构建的DRA失败分类法推导出评分指南，将失败分类为五类和十三子类。我们提出了DeepVerifier，一种基于评分指南的输出奖励验证器，利用验证和策略的不对称性，在元评估F1分数上优于普通代理-法官和LLM法官基线，提升12%-48%。为了实现实际的自我进化，DeepVerifier在测试时推理期间作为插件模块集成。验证器产生详细的评分指南反馈，反馈给代理进行迭代自举，无需额外训练即可改进响应。这种测试时扩展在使用强大闭源LLM时，在GAIA和XBench-DeepSearch的挑战子集上实现8%-11%的准确率提升。最后，为支持开源发展，我们发布了DeepVerifier-4K，一个包含4646个高质量代理步骤的curated监督微调数据集，专注于DRA验证。这些例子强调反思和自我批评，使开放模型能够发展出稳健的验证能力。
英文摘要：
Recent advances in Deep Research Agents (DRAs) are transforming automated knowledge discovery and problem-solving. While the majority of existing efforts focus on enhancing policy capabilities via post-training, we propose an alternative paradigm: self-evolving the agent's ability by iteratively verifying the policy model's outputs, guided by meticulously crafted rubrics. This approach gives rise to the inference-time scaling of verification, wherein an agent self-improves by evaluating its generated answers to produce iterative feedback and refinements. We derive the rubrics based on an automatically constructed DRA Failure Taxonomy, which systematically classifies agent failures into five major categories and thirteen sub-categories. We present DeepVerifier, a rubrics-based outcome reward verifier that leverages the asymmetry of verification and outperforms vanilla agent-as-judge and LLM judge baselines by 12%-48% in meta-evaluation F1 score. To enable practical self-evolution, DeepVerifier integrates as a plug-and-play module during test-time inference. The verifier produces detailed rubric-based feedback, which is fed back to the agent for iterative bootstrapping, refining responses without additional training. This test-time scaling delivers 8%-11% accuracy gains on challenging subsets of GAIA and XBench-DeepSearch when powered by capable closed-source LLMs. Finally, to support open-source advancement, we release DeepVerifier-4K, a curated supervised fine-tuning dataset of 4,646 high-quality agent steps focused on DRA verification. These examples emphasize reflection and self-critique, enabling open models to develop robust verification capabilities.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
