让模型自己设计适合求解者水平的推理题，提升训练数据质量。
Learning to Pose Problems: Reasoning-Driven and Solver-Adaptive Data Synthesis
- 先用推理模型生成问题设计思路，再合成带中间步骤的问题对。
- 根据求解器反馈动态调整题目难度，使其逼近其能力边界。
- 在10个基准上平均提升3.4%，适用于数学与视觉-语言模型。
为训练大型推理模型而进行的数据合成提供了一种可扩展的替代方案，克服了有限的人工标注数据瓶颈，能够生成高质量数据。然而，现有方法存在两大挑战：(i) 生成过程缺乏针对性，忽略求解器能力，产生低价值问题，或依赖复杂数据流水线来平衡难度；(ii) 问题生成缺乏深层推理，导致问题变体浅显。本文提出一种显式推理规划问题方向并自适应求解器能力的题目生成器。具体地，构建相关问题对，并通过推理模型生成中间问题设计思维链（CoT）进行增强。这些数据用于引导生成器学习问题设计策略。随后，将求解器对合成问题的反馈作为奖励信号，使生成器能校准难度，生成位于求解器能力边缘的互补性问题。在10个数学与通用推理基准上的大量实验表明，该框架实现了累计平均3.4%的性能提升，展现出在语言与视觉-语言模型中的鲁棒泛化能力。
原文摘要 · Abstract (English)
Data synthesis for training large reasoning models offers a scalable alternative to limited, human-curated datasets, enabling the creation of high-quality data. However, existing approaches face several challenges: (i) indiscriminate generation that ignores the solver's ability and yields low-value problems, or reliance on complex data pipelines to balance problem difficulty; and (ii) a lack of reasoning in problem generation, leading to shallow problem variants. In this paper, we develop a problem generator that reasons explicitly to plan problem directions before synthesis and adapts difficulty to the solver's ability. Specifically, we construct related problem pairs and augment them with intermediate problem-design CoT produced by a reasoning model. These data are used to bootstrap problem-design strategies in the generator. Then, we treat the solver's feedback on synthetic problems as a reward signal, enabling the generator to calibrate difficulty and produce complementary problems near the edge of the solver's competence. Extensive experiments on 10 mathematical and general reasoning benchmarks show that our proposed framework achieves a cumulative average improvement of 3.4%, demonstrating robust generalization across both language and vision-language models.
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。