feat: add penalize_ambiguous_claims to HallucinationMetric#2704
Open
nuthalapativarun wants to merge 1 commit into
Open
feat: add penalize_ambiguous_claims to HallucinationMetric#2704nuthalapativarun wants to merge 1 commit into
nuthalapativarun wants to merge 1 commit into
Conversation
|
@nuthalapativarun is attempting to deploy a commit to the Confident AI Team on Vercel. A member of the Team first needs to authorize it. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds a
penalize_ambiguous_claims: bool = Falseparameter toHallucinationMetric, mirroring the same flag onFaithfulnessMetric(PR #2670) andAnswerRelevancyMetric. When set toTrue, verdicts of"idk"(the LLM cannot determine agreement or contradiction) are counted as hallucinations and surfaced with an(Ambiguous)prefix in the reason.Changes
deepeval/metrics/hallucination/schema.py— added"idk"toHallucinationVerdict.verdictliteraldeepeval/metrics/hallucination/template.py— updated prompt to allow"idk"for ambiguous verdictsdeepeval/metrics/hallucination/hallucination.py— addedpenalize_ambiguous_claimsparam; updated_calculate_score,_generate_reason, and_a_generate_reasonto handle"idk"verdictstests/test_metrics/test_hallucination_metric.py— addedTestHallucinationPenalizeAmbiguousClaimswith three no-API-key unit tests that verify score calculation directlyTest plan
python -m pytest tests/test_metrics/test_hallucination_metric.py::TestHallucinationPenalizeAmbiguousClaims -v— all 3 tests pass (no API key required)python -m black --check deepeval/metrics/hallucination/— passespython -m ruff check deepeval/metrics/hallucination/— passesNotes
The change is fully backward compatible:
penalize_ambiguous_claimsdefaults toFalse, and"idk"verdicts were not possible before this change (old schema only allowed"yes"/"no"). Existing callers see no behavioral change unless they opt in.