ScaleDiff: Scaling Difficult Problems for Advanced Mathematical Reasoning

• URL: http://arxiv.org/abs/2509.21070v1
• Date: Thu, 25 Sep 2025 12:22:44 GMT
• Title: ScaleDiff: Scaling Difficult Problems for Advanced Mathematical Reasoning
• Authors: Qizhi Pei, Zhuoshi Pan, Honglin Lin, Xin Gao, Yu Li, Zinan Tang, Conghui He, Rui Yan, Lijun Wu,
• Abstract summary: Large Reasoning Models (LRMs) have shown impressive capabilities in complex problem-solving.<n>We propose ScaleDiff, a pipeline designed to scale the creation of difficult problems.<n>We show that our pipeline can effectively transfer advanced reasoning capabilities without relying on larger, more expensive teacher models.
• Score: 51.946959481392064
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large Reasoning Models (LRMs) have shown impressive capabilities in complex problem-solving, often benefiting from training on difficult mathematical problems that stimulate intricate reasoning. Recent efforts have explored automated synthesis of mathematical problems by prompting proprietary models or large-scale open-source models from seed data or inherent mathematical concepts. However, scaling up these methods remains challenging due to their high computational/API cost, complexity of prompting, and limited difficulty level of the generated problems. To overcome these limitations, we propose ScaleDiff, a simple yet effective pipeline designed to scale the creation of difficult problems. We efficiently identify difficult problems from existing datasets with only a single forward pass using an adaptive thinking model, which can perceive problem difficulty and automatically switch between "Thinking" and "NoThinking" modes. We then train a specialized difficult problem generator (DiffGen-8B) on this filtered difficult data, which can produce new difficult problems in large scale, eliminating the need for complex, per-instance prompting and its associated high API costs. Fine-tuning Qwen2.5-Math-7B-Instruct on the ScaleDiff-Math dataset yields a substantial performance increase of 11.3% compared to the original dataset and achieves a 65.9% average accuracy on AIME'24, AIME'25, HMMT-Feb'25, BRUMO'25, and MATH500, outperforming recent strong LRMs like OpenThinker3. Notably, this performance is achieved using the cost-efficient Qwen3-8B model as a teacher, demonstrating that our pipeline can effectively transfer advanced reasoning capabilities without relying on larger, more expensive teacher models. Furthermore, we observe a clear scaling phenomenon in model performance on difficult benchmarks as the quantity of difficult problems increases. Code: https://github.com/QizhiPei/ScaleDiff.

Related papers

• From Abstract to Contextual: What LLMs Still Cannot Do in Mathematics [79.81905350372067]
  We study gap through contextual mathematical reasoning.<n>We introduce ContextMATH, a benchmark that repurposes AIME and MATH-500 problems into two contextual settings.<n>Open-source models decline by 13 and 34 points on SG and CS, while proprietary models drop by 13 and 20.
  arXiv  Detail & Related papers  (2026-01-30T14:56:04Z)
• Learning to Pose Problems: Reasoning-Driven and Solver-Adaptive Data Synthesis for Large Reasoning Models [54.29243291958429]
  We develop a problem generator that reasons explicitly to plan problem directions before synthesis.<n>We treat the solver's feedback on synthetic problems as a reward signal, enabling the generator to calibrate difficulty.<n>Our method achieves an average improvement of 2.5% and generalizes to both language and vision-language models.
  arXiv  Detail & Related papers  (2025-11-13T03:08:51Z)
• LLMs Encode How Difficult Problems Are [4.990590622073335]
  We investigate whether large language models encode problem difficulty in a way that aligns with human judgment.<n>We train linear probes across layers and token positions on 60 models, evaluating on mathematical and coding subsets of Easy2HardBench.
  arXiv  Detail & Related papers  (2025-10-20T22:48:23Z)
• QueST: Incentivizing LLMs to Generate Difficult Problems [77.75835742350644]
  Large Language Models have achieved strong performance on reasoning tasks, solving competition-level coding and math problems.<n>Existing competitive coding datasets contain only thousands to tens of thousands of problems.<n>We propose QueST, a novel framework which combines difficulty-aware graph sampling and difficulty-aware rejection fine-tuning.
  arXiv  Detail & Related papers  (2025-10-20T16:29:53Z)
• Less is More Tokens: Efficient Math Reasoning via Difficulty-Aware Chain-of-Thought Distillation [82.2288581878096]
  We present a framework for difficulty-aware reasoning that teaches models to dynamically adjust reasoning depth based on problem complexity.<n>We show that models can be endowed with such dynamic inference pathways without any architectural modifications.
  arXiv  Detail & Related papers  (2025-09-05T16:40:13Z)
• SAND-Math: Using LLMs to Generate Novel, Difficult and Useful Mathematics Questions and Answers [13.763623961742391]
  We introduce textbfSAND-Math (textbfSynthetic textbfAugmented textbfNovel and textbfDifficult Mathematics problems and solutions), a pipeline that addresses high-quality problems from scratch.<n>We demonstrate the effectiveness of our approach through two key findings.
  arXiv  Detail & Related papers  (2025-07-28T05:17:48Z)
• THOUGHTTERMINATOR: Benchmarking, Calibrating, and Mitigating Overthinking in Reasoning Models [65.39456695678713]
  We introduce approximate measures of problem-level difficulty and demonstrate that a clear relationship between problem difficulty and optimal token spend exists.<n>We find that in general, reasoning models are poorly calibrated, particularly on easy problems.<n>We introduce THOUGHTTERMINATOR, a training-free black box decoding technique that significantly improves reasoning model calibration.
  arXiv  Detail & Related papers  (2025-04-17T22:16:30Z)
• Climbing the Ladder of Reasoning: What LLMs Can-and Still Can't-Solve after SFT? [59.418994222096885]
  We conduct a detailed analysis of model performance on the AIME24 dataset.<n>We categorize questions into four tiers (Easy, Medium, Hard, and Extremely Hard)<n>We find that progression from Easy to Medium tier requires adopting an R1 reasoning style with minimal SFT-1K instances.<n>Exh-level questions present a fundamentally different challenge; they require unconventional problem-solving skills.
  arXiv  Detail & Related papers  (2025-04-16T03:39:38Z)
• MATH-Perturb: Benchmarking LLMs' Math Reasoning Abilities against Hard Perturbations [90.07275414500154]
  We observe significant performance drops on MATH-P-Hard across various models.<n>We also raise concerns about a novel form of memorization where models blindly apply learned problem-solving skills.
  arXiv  Detail & Related papers  (2025-02-10T13:31:46Z)
• DART-Math: Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving [15.815363023014248]
  We propose Difficulty-Aware Rejection Tuning (DART), a method that allocates difficult queries more trials during the synthesis phase.<n>DART allocates difficult queries more trials during the synthesis phase, enabling more extensive training on difficult samples.<n>We fine-tune various base models on our datasets ranging from 7B to 70B in size, resulting in a series of strong models called DART-MATH.
  arXiv  Detail & Related papers  (2024-06-18T07:14:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.