用动作一致性自动分配计算资源，让大模型更省力地做决策。
Don't Overthink It: Inter-Rollout Action Agreement as a Free Adaptive-Compute Signal for LLM Agents
- 通过多轮输出的一致性判断任务难易，动态分配计算量。
- 在GSM8K和MiniHouse上减少超一半的调用次数，准确率不降。
- 无需训练、不依赖外部工具，适合实时推理场景。
推理时计算资源的动态扩展已成为提升大语言模型（LLM）智能体可靠性的重要手段，但现有方法对每个决策步骤均等分配计算预算。我们提出TrACE（基于轨迹一致性的自适应计算），一种无需训练的控制器，通过测量多轮输出间动作的一致性来动态分配LLM调用次数。在每一步中，TrACE采样少量候选动作并评估模型对同一动作的承诺程度：高一致性表示任务简单，立即决策；低一致性表示不确定，继续采样直至达到可配置上限后选择多数动作。无需学习组件、外部验证器或人工标注。我们在两个基准测试上评估TrACE，涵盖单步推理（GSM8K，n=50）和多步家庭导航（MiniHouse，n=30），使用运行于CPU的Qwen 2.5 3B Instruct模型。TrACE-4在GSM8K上达到与SC-4相当的准确率，但减少33%的调用；在MiniHouse上减少39%。TrACE-8在GSM8K上匹配SC-8准确率，调用减少55%；在MiniHouse上减少65%。进一步验证了跨轮次动作一致性是任务成功率的可靠信号，证明模型自身输出一致性可编码难度信息，无需训练即可利用。TrACE是首个在多步序列决策任务上评估的无训练、逐时刻自适应计算控制器。
原文摘要 · Abstract (English)
Inference-time compute scaling has emerged as a powerful technique for improving the reliability of large language model (LLM) agents, but existing methods apply compute uniformly: every decision step receives the same budget regardless of its difficulty. We introduce TrACE (Trajectorical Adaptive Compute via agrEement), a training-free controller that allocates LLM calls adaptively across agent timesteps by measuring inter-rollout action agreement. At each step, TrACE samples a small set of candidate next actions and measures how consistently the model commits to the same action. High agreement signals an easy decision; the controller commits immediately. Low agreement signals uncertainty; the controller samples additional rollouts up to a configurable cap before committing to the plurality action. No learned components, no external verifier, and no human labels are required. We evaluate TrACE against greedy decoding and fixed-budget self-consistency (SC-4, SC-8) on two benchmarks spanning single-step reasoning (GSM8K, n=50) and multi-step household navigation (MiniHouse, n=30), using a Qwen 2.5 3B Instruct model running on CPU. TrACE-4 matches SC-4 accuracy while using 33% fewer LLM calls on GSM8K and 39% fewer on MiniHouse. TrACE-8 matches SC-8 accuracy with 55% fewer calls on GSM8K and 65% fewer on MiniHouse. We further show that inter-rollout agreement is a reliable signal of step-level success, validating the core hypothesis that the model's own output consistency encodes difficulty information that can be exploited without training. TrACE is the first training-free, per-timestep adaptive-compute controller for LLM agents to be evaluated on multi-step sequential decision tasks.
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。