让信息搜索智能体并行思考，提升效率与答案质量。
ParallelMuse: Agentic Parallel Thinking for Deep Information Seeking
- 分段规划路径，按不确定性复用和分支探索，提高效率
- 压缩推理过程，保留关键信息，生成连贯最终答案
- 在多个基准上性能提升62%，探索耗能降低10%~30%
并行思考拓展了探索广度，弥补了信息搜索（IS）智能体深度探索的不足，进一步增强问题解决能力。然而，传统并行思考在此场景面临两大挑战：从头重复展开导致效率低下，以及长时推理轨迹难以整合，因上下文容量有限无法充分考虑推理过程。为此，我们提出ParallelMuse，一种专为深度信息搜索智能体设计的两阶段范式。第一阶段，功能指定部分回溯，将生成序列划分为功能区域，通过不确定性引导的路径复用与分支，提升探索效率；第二阶段，压缩推理聚合，利用推理冗余实现对答案推导相关的信息无损压缩，并合成连贯的最终答案。在多个开源智能体和基准上的实验表明，该方法可实现最高62%的性能提升，同时探索令牌消耗减少10%至30%。
原文摘要 · Abstract (English)
Parallel thinking expands exploration breadth, complementing the deep exploration of information-seeking (IS) agents to further enhance problem-solving capability. However, conventional parallel thinking faces two key challenges in this setting: inefficiency from repeatedly rolling out from scratch, and difficulty in integrating long-horizon reasoning trajectories during answer generation, as limited context capacity prevents full consideration of the reasoning process. To address these issues, we propose ParallelMuse, a two-stage paradigm designed for deep IS agents. The first stage, Functionality-Specified Partial Rollout, partitions generated sequences into functional regions and performs uncertainty-guided path reuse and branching to enhance exploration efficiency. The second stage, Compressed Reasoning Aggregation, exploits reasoning redundancy to losslessly compress information relevant to answer derivation and synthesize a coherent final answer. Experiments across multiple open-source agents and benchmarks demonstrate up to 62% performance improvement with a 10--30% reduction in exploratory token consumption.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。