Fix prompts for audiollm for misrepresenting_tool_result and information_loss#187
Merged
Merged
Conversation
…pt for faithfulness for audiollm vs s2s
katstankiewicz
changed the title
fanny-riols
reviewed
Jul 16, 2026
| carve-outs, the assistant-turns disclaimer) must gate on this, not on | ||
| ``is_audio_native``. | ||
| """ | ||
| return self.pipeline_type == PipelineType.S2S |
Collaborator
There was a problem hiding this comment.
That won't match ElevenLabs agents that is marked as s2s, right?
Collaborator
Author
There was a problem hiding this comment.
No, we switch elevenlabs agents pipeline type to cascade earlier in the process so it is cascade when it reaches here
gabegma
reviewed
Jul 16, 2026
| assert "STT transcriptions of the agent" not in variables["assistant_turns_disclaimer"] | ||
| assert "intended text" in variables["assistant_turns_disclaimer"] | ||
|
|
||
| def test_disambiguation_context_drops_comparative_bar(self): |
Collaborator
There was a problem hiding this comment.
This test is a bit weird to me - I don't think we need to test its not longer there.
Collaborator
Author
There was a problem hiding this comment.
yeah good point! i'll remove
gabegma
approved these changes
Jul 16, 2026
gabegma
left a comment
gabegma
left a comment
Collaborator
There was a problem hiding this comment.
Excellent, thank you!! I wanted to do the change from audio_native to s2s for some time now for those 2 metrics for the assistant turns only, this is perfect 👌
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
information_lossandmisrepresenting_tool_resultimplied that the text was a transcript for audio llm so to not judge if mistakes are made, but this is inaccurate for audio llms since it is the actual output. If the output is incorrect assume it is a model issue