Skip to content

research_mining

research_mining

Research mining scorer — LLM judge for research synthesis quality.

Evaluates accuracy, depth, source quality, and synthesis of AI-generated research responses.

Classes

ResearchMiningScorer 

ResearchMiningScorer(
    judge_backend: InferenceBackend, judge_model: str
)

Bases: LLMJudgeScorer

Score research mining responses on quality dimensions.

Source code in src/diapason/evals/core/scorer.py 
33
34
35
def __init__(self, judge_backend: InferenceBackend, judge_model: str) -> None:
    self._judge_backend = judge_backend
    self._judge_model = judge_model