RE-TRAC: 递归轨迹压缩用于深度搜索代理
RE-TRAC: REcursive TRAjectory Compression for Deep Search Agents
AI总结：
Re-TRAC通过递归轨迹压缩提升深度搜索代理的效率，实现跨轨迹探索与全局规划，显著优于ReAct框架。
摘要：
基于大语言模型的深度研究代理大多基于ReAct框架。这种线性设计使得难以回溯早期状态、分支到替代的搜索方向或在长上下文中维持全局意识，常常导致局部最优、冗余探索和低效搜索。我们提出了Re-TRAC，一种代理框架，通过在每次轨迹后生成结构化的状态表示来实现跨轨迹探索，以总结证据、不确定性、失败和未来计划，并基于此状态表示来条件后续轨迹。这使得能够进行迭代反思和全局知情规划，将研究重新定义为一个渐进的过程。实证结果表明，Re-TRAC在前沿大语言模型上的BrowseComp任务中，比ReAct高出15-20%。对于较小的模型，我们引入了Re-TRAC-aware监督微调，实现了在可比规模下的最佳性能。值得注意的是，Re-TRAC在各轮次中工具调用和令牌使用量呈现单调减少，表明由跨轨迹反思驱动的逐步定向探索，而非冗余搜索。
英文摘要：
LLM-based deep research agents are largely built on the ReAct framework. This linear design makes it difficult to revisit earlier states, branch into alternative search directions, or maintain global awareness under long contexts, often leading to local optima, redundant exploration, and inefficient search. We propose Re-TRAC, an agentic framework that performs cross-trajectory exploration by generating a structured state representation after each trajectory to summarize evidence, uncertainties, failures, and future plans, and conditioning subsequent trajectories on this state representation. This enables iterative reflection and globally informed planning, reframing research as a progressive process. Empirical results show that Re-TRAC consistently outperforms ReAct by 15-20% on BrowseComp with frontier LLMs. For smaller models, we introduce Re-TRAC-aware supervised fine-tuning, achieving state-of-the-art performance at comparable scales. Notably, Re-TRAC shows a monotonic reduction in tool calls and token usage across rounds, indicating progressively targeted exploration driven by cross-trajectory reflection rather than redundant search.
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。
