📋 Welcome to Rubric Library

Discover our curated collection of evaluation rubrics designed to standardize and enhance response quality assessment. Our library includes both query-agnostic rubrics for domain-specific evaluation and query-specific rubrics for targeted assessment scenarios.

🎯 What you'll find:

• Query-Agnostic Rubrics: Domain-specific evaluation criteria for general, code, math, science, technology, and engineering fields
• Query-Specific Rubrics: Targeted assessment frameworks tailored to specific user queries and scenarios. Query-specific rubrics are provided in this dataset: https://huggingface.co/datasets/agentscope-ai/Auto-Rubric, which includes detailed evaluation criteria tailored to individual queries.
• Multi-Domain Coverage: From Python code quality to physics problem-solving, cybersecurity to system design
• Complexity Levels: Rubrics ranging from medium to very high complexity for various assessment needs
• Source Diversity: Community standards, academic frameworks, and industry best practices

💡 Use the search bar below to find specific rubrics, or browse by category and domain to explore our full collection.