OpenRubrics/RubricARROW-8B-Rubric

RubricARROW-8B-Rubric: Specialized Rubric Generation Model

RubricARROW-8B-Rubric is an 8 billion parameter language model, fine-tuned from the Qwen3-8B architecture. Its core function is to automatically generate detailed, rubric-style evaluation criteria based on a user's request. This model is particularly adept at transforming specific instructions into universal principles for assessment.

Key Capabilities

• Rubric Extraction: Extracts a set of rubric-style instructions from natural language requests.
• Categorization: Distinguishes between "Hard Rule" rubrics (derived from explicit requirements like format or length) and "Principle" rubrics (abstracted, domain-agnostic quality criteria such as clarity or correctness).
• Universality: Ensures all rubric items are universal principles, free from topic-specific references, names, or numbers.
• Comprehensiveness: Generates rubrics that cover all critical aspects, including explicit requirements and implicit quality standards.
• Conciseness & Uniqueness: Produces distinct evaluation criteria, merging redundant items and using precise, repetition-free wording.
• Structured Output: Formats rubrics as a numbered list, with each item starting "The response" and appended with [Hard Rule] or [Principle].

Good For

• Automating the creation of evaluation rubrics for various tasks.
• Standardizing assessment criteria in educational or quality assurance contexts.
• Generating objective guidelines for evaluating LLM responses or other creative outputs.
• Applications requiring structured, principle-based feedback mechanisms.