为大模型推理设计动态选择策略，提升任务表现。
Select-then-Solve: Paradigm Routing as Inference-Time Optimization for LLM Agents
- 用轻量嵌入路由选择最佳推理范式，动态适配任务
- 平均准确率从47.6%提升至53.1%，超最优固定范式2.8个百分点
- 适合需要灵活推理的复杂任务，尤其对弱模型效果显著
当基于大模型的智能体在任务上表现提升时，这种进步是来自模型本身还是推理范式？我们通过对比六种推理范式（直接、思维链、ReAct、计划执行、反思、ReCode）在四个前沿大模型和十个基准上的表现，共完成约18,000次实验。结果发现：某些任务中推理结构显著提升性能（如ReAct在GAIA上比直接法高44个百分点），而在另一些任务中反而降低（如CoT在HumanEval上下降15个百分点）。没有单一范式能通用于所有任务；针对每项任务进行最优选择可使平均性能优于最优固定范式17.1个百分点。为此，我们提出“先选择后求解”方法：在回答每个任务前，由轻量级嵌入路由选择最合适的范式。该方法在四个模型上将平均准确率从47.6%提升至53.1%，超越最优固定范式（50.3%）2.8个百分点，并恢复了高达37%的“理想选择差距”。相比之下，零样本自路由仅在GPT-5上达到67.1%的表现，其他模型均失败，远不及学习型路由。结果表明，推理范式的选择应作为每任务的动态决策，而非固定架构设计。
原文摘要 · Abstract (English)
When an LLM-based agent improves on a task, is the gain from the model itself or from the reasoning paradigm wrapped around it? We study this question by comparing six inference-time paradigms, namely Direct, CoT, ReAct, Plan-Execute, Reflection, and ReCode, across four frontier LLMs and ten benchmarks, yielding roughly 18,000 runs. We find that reasoning structure helps dramatically on some tasks but hurts on others: ReAct improves over Direct by 44pp on GAIA, while CoT degrades performance by 15pp on HumanEval. No single paradigm dominates, and oracle per-task selection beats the best fixed paradigm by 17.1pp on average. Motivated by this complementarity, we propose a select-then-solve approach: before answering each task, a lightweight embedding-based router selects the most suitable paradigm. Across four models, the router improves average accuracy from 47.6% to 53.1%, outperforming the best fixed paradigm at 50.3% by 2.8pp and recovering up to 37% of the oracle gap. In contrast, zero-shot self-routing only works for GPT-5 at 67.1% and fails for weaker models, all trailing the learned router. Our results argue that reasoning paradigm selection should be a per-task decision made by a learned router, not a fixed architectural choice.
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。