agentlans/Qwen2.5-1.5B-Instruct-Multiple-Choice-Maker
The agentlans/Qwen2.5-1.5B-Instruct-Multiple-Choice-Maker is a 1.5 billion parameter instruction-tuned causal language model, based on the Qwen2.5 architecture, developed by agentlans. It is specifically fine-tuned to generate multiple-choice questions in XML format from input text. This model excels at creating assessment materials for educational purposes and generating benchmark questions for machine learning evaluations, offering a specialized solution for automated question generation.
Loading preview...
Overview
This model, agentlans/Qwen2.5-1.5B-Instruct-Multiple-Choice-Maker, is a 1.5 billion parameter Qwen2.5-based instruction-tuned model specifically designed for generating multiple-choice questions in XML format. It is fine-tuned to process input text and produce structured quiz content, making it a specialized tool for educational and machine learning applications.
Key Capabilities
- XML-formatted Question Generation: Produces multiple-choice questions with choices and correct answers embedded within a
<quiz>XML structure. - Educational Content Creation: Automates the generation of assessment materials for exams and quizzes.
- Machine Learning Benchmark Generation: Useful for creating datasets to evaluate natural language processing models.
- Specialized Fine-tuning: Leverages a custom dataset derived from open-source sociology textbooks and the
agentlans/finewebedu-multiple-choicedataset, fine-tuned using LLaMA-Factory with LoRA.
Good for
- Educators and Content Creators: Streamlining the process of creating multiple-choice questions from textual content.
- Researchers and Developers: Generating structured benchmark datasets for evaluating NLP models' understanding and question-answering capabilities.
- Automated Assessment Systems: Integrating into systems that require programmatic generation of quiz content.
Limitations
- Generated outputs may require manual review to ensure contextual accuracy, appropriateness, and that only one correct answer is provided.
- The XML output, while robust for validation, requires further processing for human readability or direct machine assessment.