EviGraph：面向信息检索智能体的可验证证据构建
EviGraph: Towards Verifiable Evidence Construction for Information-Seeking Agents
浏览论文内容
中文总结 AI 辅助
EviGraph是分离搜索与证据记录的深度搜索框架，在BrowseComp等数据集上显著提升了智能体网络搜索的准确率，每轮生成token更少，验证了结构化证据记录的有效性。
中文摘要 AI 辅助
智能体网络搜索可检索相关信息，但无法证明检索内容实际支持答案所用的主张。现有智能体通常将搜索与证据记录保存在线性交互轨迹中，主要针对最终答案的正确性优化，对中间基础的监督有限。我们提出EviGraph，一种深度搜索框架，该框架将搜索执行与证据记录分离，同时为可训练角色使用共享策略：执行器规划简洁查询，冻结的证据验证器检查源页面并返回带有明确极性的逐字证据项，策略映射这些项以添加/支持图请求，由确定性结构验证器检查。生成的图既作为持久工作记忆，又作为密集过程奖励的来源，使强化学习能直接监督证据构建，而非仅监督最终答案。在BrowseComp-Plus上，Qwen3-8B EviGraph智能体在匹配的交互预算下达到35.9%的准确率，相比之下，相同双角色架构无强化学习的智能体准确率为26.9%，整体式智能体仅为2.7%，且每轮生成的token更少。在BrowseComp、GAIA和XBench上的一致提升表明，显式构建证据记录结构并对其进行奖励可改进智能体搜索。
英文摘要
Agentic Web search can retrieve relevant information without establishing that the retrieved content actually supports the claims used in an answer. Existing agents typically keep search and evidence recording in a linear interaction trace and optimize primarily for final-answer correctness, providing limited supervision for intermediate grounding. We present EviGraph, a deep-search framework that separates search execution from evidence recording while using a shared policy for the trainable roles. An executor plans concise queries, a frozen evidence verifier inspects source pages and returns verbatim evidence items with an explicit polarity, and the policy maps those items to add/support graph requests that are checked by a deterministic structural validator. The resulting graph serves both as persistent working memory and as a source of dense process rewards, enabling reinforcement learning to directly supervise evidence construction rather than only the final answer. On BrowseComp-Plus, a Qwen3-8B EviGraph agent achieves 35.9% accuracy under a matched interaction budget, compared with 26.9% for the same dual-role architecture without reinforcement learning and 2.7% for a monolithic agent, while generating fewer tokens per rollout. Consistent gains on BrowseComp, GAIA, and XBench indicate that explicitly structuring and rewarding evidence recording improves agentic search
