DeepRewind：预测并修复深度研究智能体中的过早承诺
DeepRewind: Predicting and Repairing Premature Commitments in Deep Research Agents
- University of British Columbia（不列颠哥伦比亚大学）
机构由 AI 辅助整理，请以论文原文为准。
中文总结 AI 辅助
针对深度研究智能体过早承诺问题，提出DeepRewind附加控制层，通过类型化图表示认知状态、预测可逆性并阻止风险承诺，在基准上将洞察召回率提升3.6个百分点，过早承诺减少59.1%。
中文摘要 AI 辅助
深度研究智能体通过迭代搜索、证据评估、信念修正和综合归纳来开展长期调查。然而，它们可能在获得充分证据之前就对某些主张做出承诺，导致后续推理强化错误的解读。我们提出了DeepRewind，一种用于可逆深度研究的附加控制层，它将智能体不断演化的认知状态表示为包含来源、证据、主张、假设、假设、承诺、计划和草稿的类型化图。在接受中间结论之前，一个基于提示的世界模型会预测其影响，并基于假设收窄、信息损失、恢复成本和矛盾触发覆盖来估计可逆性。一个二元控制器会阻止有风险的承诺，而一致性监视器会在后续证据使这些承诺失效时执行依赖感知的回滚。在DRBench和LiveDRBench上，DeepRewind将洞察召回率提高了3.6个百分点，并将过早承诺相对于Open Deep Research减少了59.1%。
英文摘要
Deep-research agents conduct long-horizon investigations through iterative search, evidence evaluation, belief revision, and synthesis. However, they may commit to claims before sufficient evidence is available, causing later reasoning to reinforce an incorrect interpretation. We introduce DeepRewind, an additive control layer for reversible deep research that represents the agent's evolving epistemic state as a typed graph of sources, evidence, claims, hypotheses, assumptions, commitments, plans, and drafts. Before accepting an intermediate conclusion, a prompt-based world model predicts its impact and estimates reversibility based on hypothesis narrowing, information loss, recovery cost, and contradiction-trigger coverage. A binary controller blocks risky commitments, while a consistency monitor performs dependency-aware rollback when later evidence invalidates them. Across DRBench and LiveDRBench, DeepRewind improves insight recall by 3.6 percentage points and reduces premature commitments by 59.1% relative to Open Deep Research.
