/rdagent/scenarios/finetune/datasets/deepscaler

language, size_categories, license, configs

language	size_categories	license	configs
en	10K<n<100K	mit	config_name data_files splits default split path train data/train-* name num_examples train 40315

DeepScaleR Mathematical Reasoning Dataset

Dataset for DeepScaleR: Surpassing O1-Preview with a 1.5B Model by Scaling RL.

DeepScaleR-1.5B-Preview achieves 43.1% Pass@1 accuracy on AIME 2024, representing a 15% improvement over the base model (28.8%) and surpassing OpenAI's O1-Preview performance with just 1.5B parameters through distributed reinforcement learning.

Overview

The DeepScaleR dataset is a carefully curated collection of approximately 40,000 unique mathematics problem-answer pairs designed for training mathematical reasoning models through reinforcement learning. This dataset covers a wide range of competition-level mathematics problems from high school to olympiad level, providing a robust foundation for scaling RL algorithms on reasoning tasks.

DeepScaleR demonstrates that sophisticated mathematical reasoning can be achieved through strategic data curation combined with iterative context length scaling (8K→16K→24K) using Group Relative Policy Optimization (GRPO).

Data Sources

Our training dataset consists of problems compiled from prestigious mathematics competitions and curated datasets:

• AIME (American Invitational Mathematics Examination) problems (1984-2023)
• AMC (American Mathematics Competition) problems (prior to 2023)
• Omni-MATH dataset
• Still dataset

Data Fields

The dataset contains three key fields:

• problem: The mathematical problem statement, formatted with LaTeX notation
• solution: Official solution to the problem, including LaTeX formatting and boxed final answers. If there is no solution, the solution field is an empty string
• answer: The final mathematical result/answer, usually extracted from the solution

CoT Quality Assessment

IMPORTANT: Raw data must be polished before training.

Dimension	Value
baseline_quality	low
task_type	math
polish_difficulty	high

Baseline: 82% empty solution, 18% too short (p50=373 tokens, summary-style). Need to generate exploratory CoT (For your reference, the length of a well-structured CoT is usually longer than 1/4 * the model max_position_embeddings tokens) for all samples.

License

This dataset is released under the MIT License.