← 研踪 PaperTrailarXiv:2611.06662 · v1
Where Search Agents Go Off the Rails: Graph-Grounded Step-Level Verification
首次提交 2026-11-04 23:37
版本更新 2026-11-04 23:37
雷达观测 2026-11-05 12:13
摘要
Search-augmented reasoning systems can reach correct answers while making errors in intermediate steps. Reliable auditing must check whether evidence $D$ covers the question $Q$, reasoning $T$ addresses $Q$, and $T$ is grounded in $D$. We introduce GraphRail, which represents requirements, evidence, and reasoning as structured claims and compares them without a generative judge or gold answer. Across three multi-hop QA benchmarks and three reasoning frameworks, GraphRail improves error recall over generative verifiers while remaining stable under stochastic extraction. Its diagnoses can also guide online reasoning, providing a traceable way to detect and correct intermediate errors.
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-04 23:37 · 最新观测版本
Where Search Agents Go Off the Rails: Graph-Grounded Step-Level Verification
Search-augmented reasoning systems can reach correct answers while making errors in intermediate steps. Reliable auditing must check whether evidence $D$ covers the question $Q$, reasoning $T$ addresses $Q$, and $T$ is grounded in $D$. We introduce GraphRail, which represents requirements, evidence, and reasoning as structured claims and compares them without a generative judge or gold answer. Across three multi-hop QA benchmarks and three reasoning frameworks, GraphRail improves error recall over generative verifiers while remaining stable under stochastic extraction. Its diagnoses can also guide online reasoning, providing a traceable way to detect and correct intermediate errors.
查看此版本解读 →查看此版本原文 ↗