π Irrelevant InformationΒΆ
metrics.llm_as_judge.direct.criteria.irrelevant_information
CriteriaWithOptions(
name="irrelevant_information",
description="Does the user response contain irrelevant information?",
options=[
CriteriaOption(
name="Yes",
description="The user response contains irrelevant information.",
),
CriteriaOption(
name="No",
description="The user response doesn't contain irrelevant information.",
),
],
option_map={
"Yes": 0.0,
"No": 1.0,
},
)
[source]from unitxt.llm_as_judge_constants import CriteriaOption
Read more about catalog usage here.