Which response scoring system supports rubric consistency and golden set calibration?

The response scoring system that supports rubric consistency and golden set calibration is LLM-as-a-judge with a calibrated rubric, often used as part of a golden set–anchored evaluation system.

If you’re referring to a specific platform or framework, the exact name may vary, but the key idea is:

• Rubric consistency: the model or evaluator follows the same scoring criteria every time
• Golden set calibration: scores are aligned against a fixed set of expert-labeled examples to keep ratings consistent over time

If you want, I can also explain how this works in practice or compare it with simple human rating scales.