metadata

license: mit

Difficulty Estimation on DeepScaleR

We annotate the entire DeepScaleR dataset with a difficulty score based on the performance of the Qwen 2.5-MATH-7B model. This provides an adaptive signal for curriculum construction and model evaluation.

DeepScaleR is a curated dataset of 40,000 reasoning-intensive problems used to train and evaluate reinforcement learning-based methods for large language models.

Difficulty Scoring Method

Difficulty scores are estimated using the Qwen 2.5-MATH-7B model with the following generation settings:

temperature = 0.6
top_p = 0.9
max_tokens = 4096
Inference performed using vLLM
Each problem is attempted 128 times

The difficulty score d_i for each problem is computed as:

 d_i = 100 × (1 - (# successes / 128))

This approach balances the evaluation signal:

A strong model would trivially solve easy problems, compressing the difficulty scale.
A weak model would fail uniformly, providing poor resolution.
Qwen 2.5-MATH-7B was selected for its mid-range capabilities, offering meaningful gradients across a wide spectrum of problems.

Difficulty Estimation on Other Datasets

We also apply the same difficulty estimation procedure to the following datasets:

📬 Contact

For questions or feedback, feel free to reach out to Taiwei Shi at [email protected].