MathV360K: 360K Multimodal Math Question-Answer Pairs

MathV360K is a multimodal dataset containing 360,000 question-answer pairs and 40,000 images sourced from 24 datasets. It was created by Zhiqiang007 and uploaded to Hugging Face on 2024-06-27 to enhance the mathematical reasoning capabilities of multimodal large language models.

Use Cases

• Fine-tuning MLLMs for mathematical problem-solving based on the described 360K question-answer pairs.
• Benchmarking model performance on multimodal math tasks based on the described use in the MathVista and MathVision benchmarks.
• Training models to interpret mathematical diagrams and text based on the described combination of 40K images and textual questions.

Strengths

• Contains 360,000 question-answer pairs, providing a substantial volume of training data.
• Integrates 40,000 images from 24 different source datasets, suggesting diversity in visual mathematical content.
• Was used to achieve a reported 46.6% accuracy on the MathVista benchmark.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Freshness should be verified as the last update was 2024-06-27.

Provenance

Source

Hugging Face, aggregated from 24 datasets.

Collection Method

Bootstrapped for the Math-LLaVA project.

Freshness

Last updated 2024-06-27 10:23:19