πŸ“„ Judge With Question Simplified NumericΒΆ

templates.rag_eval.faithfulness.judge_with_question_simplified_numeric

type: InputOutputTemplateWithCustomTarget
input_format: "Question: {question}\n\nEvidence: {contexts}\n\nPrediction: {answer}\n"
output_format: {is_faithful}
postprocessors: 
  - processors.take_first_word
  - processors.lower_case
  - processors.cast_to_float_return_zero_if_failed
  - processors.scale_0_10_to_0_1
reference: {number_val}
target_prefix: Answer: 
instruction: "You are given a question, the corresponding evidence and a prediction from a model. Compare the \"Prediction\" and the \"Evidence\" to determine to what extent the prediction is grounded in the evidence.\nTo be grounded in the evidence, all the information of the prediction must either be present in the evidence or deducible from the evidence.\n\nThe question is only given for context, and is irrelevant for determining the groundedness of the prediction.\nOn a scale of 0 to 10, to what extent is the prediction grounded in the evidence? Reply with your rating score without any preceding explanation.\n\n"
[source]

References: processors.cast_to_float_return_zero_if_failed, processors.scale_0_10_to_0_1, processors.take_first_word, processors.lower_case

Read more about catalog usage here.