根据问题自动选择推理深度，让大模型更高效解题。
Ada-R1: Hybrid-CoT via Bi-Level Adaptive Reasoning Optimization
- 混合长/短思维链模型，支持多种推理风格。
- 在五大数据集上推理长度平均减少50%以上。
- 适合需要高效推理的数学与逻辑任务场景。
近期，长思维链（Long-CoT）推理模型在复杂任务中表现优异，但推理开销大，效率成瓶颈。实证分析发现，长思维链的效果因问题而异：部分问题需深入推理，另一些则无提升甚至准确率下降。这促使发展自适应推理策略，按输入动态调整推理深度。然而，现有方法多聚焦于消除长推理路径中的冗余，未能突破长思维链范式。为此，我们提出一种两阶段自适应高效推理框架：首先融合长、短思维链模型构建混合推理模型，实现多样化推理风格；其次采用双层偏好训练，引导模型在组级别选择合适推理风格，同时在组内偏好简洁且正确的推理路径。实验表明，所提方法Ada-R1显著降低推理成本，同时保持性能。在五个数学数据集上，平均推理长度减少超过50%，凸显自适应策略优化大模型推理效率的潜力。代码即将发布于 https://github.com/StarDewXXX/AdaR1。
原文摘要 · Abstract (English)
Recently, long-thought reasoning models achieve strong performance on complex reasoning tasks, but often incur substantial inference overhead, making efficiency a critical concern. Our empirical analysis reveals that the benefit of using Long-CoT varies across problems: while some problems require elaborate reasoning, others show no improvement, or even degraded accuracy. This motivates adaptive reasoning strategies that tailor reasoning depth to the input. However, prior work primarily reduces redundancy within long reasoning paths, limiting exploration of more efficient strategies beyond the Long-CoT paradigm. To address this, we propose a novel two-stage framework for adaptive and efficient reasoning. First, we construct a hybrid reasoning model by merging long and short CoT models to enable diverse reasoning styles. Second, we apply bi-level preference training to guide the model to select suitable reasoning styles (group-level), and prefer concise and correct reasoning within each style group (instance-level). Experiments demonstrate that our method (Ada-R1) significantly reduces inference costs compared to other baseline approaches, while maintaining performance. Notably, on five mathematical datasets, the average length of reasoning is reduced by more than 50%, highlighting the potential of adaptive strategies to optimize reasoning efficiency in large language models. Our code is coming soon at https://github.com/StarDewXXX/AdaR1
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。