Conversation
Collaborator
Author
|
With some more filtering of samples for audio quality, the dataset now has 147 samples. |
updating turn handling for multi-turn evals
#22) * added phonetics, speech_disorder, and speech_enhancement tasks - still in need of full model scoring. Fixed small inconsistency bug in config by changing judge_properties to judge_settings. * Update the correct HF path for noise_detection task * updated scores --------- Co-authored-by: hoang <huuhoang.nguyen@servicenow.com>
… into scratch/gpqa
nhhoang96
approved these changes
Apr 18, 2026
Collaborator
nhhoang96
left a comment
nhhoang96
left a comment
There was a problem hiding this comment.
LGTM
Resolving documentation conflicts before merging to main
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
7 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
📌 Description
Adds GPQA Diamond Audio.
Has 155 of 198 speakable samples converted to speech for evaluation at ServiceNow-AI/gpqa_audio.
On parallel text run, GPT 4o mini scores 39 (reported 40.8 at https://openai.com/index/gpt-4o-mini-advancing-cost-efficient-intelligence/)
On audio:
GPT 4o mini: 28.9 +- 0.86 (5 runs)
Voxtral Small: 27.1
Phi 4 Multimodal Instruct: 22.58
🛠️ Type of Change
✅ How Has This Been Tested?
Test Results / Screenshots (if applicable):
📸 Screenshots / Demos
📋 Checklist
🙌 Additional Notes