Skip to content

[MVEB] Add Daily-Omni video-centric QA task#4530

Merged
isaac-chung merged 2 commits intoembeddings-benchmark:mainfrom
Rakshitha-Ireddi:mveb-dailyomni-vcqa
Apr 30, 2026
Merged

[MVEB] Add Daily-Omni video-centric QA task#4530
isaac-chung merged 2 commits intoembeddings-benchmark:mainfrom
Rakshitha-Ireddi:mveb-dailyomni-vcqa

Conversation

@Rakshitha-Ireddi
Copy link
Copy Markdown
Contributor

Adds DailyOmniVideoCentricQA — a multiple-choice video QA task following the
AbsTaskRetrieval + RetrievalSplitData pattern established in NExT-QA (#4462).

Dataset: mteb/daily-omni-video-centric-qa
Category: vt2t
Modalities: video, audio, text
Metric: accuracy

Baseline results

Random encoder baseline: 26.5% accuracy (test split)
DailyOmniVideoCentricQA.json

@Rakshitha-Ireddi Rakshitha-Ireddi mentioned this pull request Apr 27, 2026
72 tasks
Copy link
Copy Markdown
Collaborator

@isaac-chung isaac-chung left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I have the exact same comments as in #4525

Comment thread mteb/tasks/multichoice/eng/daily_omni.py Outdated
@Samoed Samoed added new dataset Issues related to adding a new task or dataset video video extension labels Apr 28, 2026
@Rakshitha-Ireddi
Copy link
Copy Markdown
Contributor Author

Hello @isaac-chung , can you check this one as well

@Samoed
Copy link
Copy Markdown
Member

Samoed commented Apr 29, 2026

@Rakshitha-Ireddi Please, don't force push. This is hard to review

Copy link
Copy Markdown
Member

@Samoed Samoed left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You need to update __init__ to make possibe to import your tasks

@isaac-chung isaac-chung merged commit 285e68c into embeddings-benchmark:main Apr 30, 2026
13 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

new dataset Issues related to adding a new task or dataset video video extension

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants