Chinese Mining Engineering Instruction Dataset with 5,287 Q&A Pairs

5,287 Chinese question-answer pairs were created by Hefei University of Technology students to enhance LLMs in mining engineering. The dataset covers six modules including laws, specifications, and safety accident cases. It was last updated on July 24, 2025, and includes a version enhanced with chain-of-thought reasoning.

Use Cases

• Instruction tuning of LLMs based on domain-specific Q&A pairs.
• Evaluating model reasoning on safety case studies mentioned in the description.
• Benchmarking model knowledge of engineering regulations and professional terminology.
• Training models to generate chain-of-thought explanations for technical problems.

Strengths

• Contains 5,287 high-quality Chinese Q&A pairs for supervised fine-tuning.
• Covers six distinct knowledge modules including laws, specifications, and safety cases.
• Includes a separate version with chain-of-thought annotations to enhance reasoning.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the evaluation and CoT-enhanced sets is unknown, which may limit suitability assessment.
• Data may reflect the academic and project-specific bias inherent to its creation by university students.

Provenance

Source

Hefei University of Technology student project (acnul).

Collection Method

Constructed as part of a university innovation training program.

Freshness

Last updated 2025-07-24 00:02:23; freshness should be verified.