让AI代理智能省略冗余推理，提升效率。
Adaptive Latent Agentic Reasoning
- 用隐变量代替冗长文本推理，只在关键步骤才展开思考。
- 搜索任务减少43.6%生成文本，工具使用减少84.6%。
- 适合追求高效推理的AI代理开发与部署场景。
大型推理模型通过生成扩展的思维链（CoT）提升性能，但在应用于大语言模型代理时效率下降。现有代理在每一步决策中常生成冗长的文本推理，且推理投入几乎均匀分布，导致多轮代理轨迹效率低下。本文提出自适应隐式代理推理（ALAR），一种双模式框架：常规步骤使用紧凑隐式推理，仅在需要深入思考时才切换为显式思维链。ALAR通过代理行为作为监督锚点学习隐式推理，并优化策略，在任务成功可由隐式推理完成时优先使用，仅对难题保留显式思维链。在代理搜索与工具使用基准测试中，ALAR在保持或提升任务准确率的同时，搜索任务生成文本量减少43.6%，工具使用减少84.6%。结果表明，ALAR通过减少不必要的文本推理，在不牺牲深度思考的前提下显著改善了大模型代理的准确性-效率权衡。
原文摘要 · Abstract (English)
Large reasoning models improve performance by generating extended chain-of-thought (CoT) reasoning, but this behavior becomes inefficient when applied to LLM agents. Current LLM agents often generate verbose textual reasoning at every decision step and allocate reasoning effort nearly uniformly across turns, leading to substantial inefficiency in multi-turn agentic trajectories. We propose Adaptive Latent Agentic Reasoning (ALAR), a dual-mode framework that uses compact latent reasoning for routine turns and selectively escalates to explicit chain-of-thought when deeper deliberation is needed. ALAR learns latent reasoning by using the agent's actions as supervision anchors and is further optimized to use latent reasoning when it is sufficient for task success and reserve explicit CoT for harder decisions. Experiments on agentic search and tool-use benchmarks show that ALAR maintains comparable or better task accuracy while substantially reducing generated tokens by up to 43.6% in search and 84.6% in tool use. These results demonstrate that ALAR improves the accuracy-efficiency trade-off of LLM agents by reducing unnecessary textual reasoning while preserving explicit deliberation for harder decision steps.
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。