Hi,
Niels here from the open-source team at Hugging Face. Congratulations on your work!
I've made the paper and 6 paper-native evaluations available on Papers with Code.
The paper has results on Image Understanding and Video Understanding task pages.
The GIT2 result currently ranks first on COCO Captions Image-to-Text Retrieval.
The GIT2 results currently rank fourth on MSVD-QA, NoCaps, and TextCaps.
Would it be possible to verify these results and let me know if any score, model name, benchmark protocol, or openness metadata should be corrected?
You can also edit the task, methods, project page, and GitHub URL directly from the paper page using your Hugging Face account.
If you'd like to showcase the results in your repository README, you can copy these live leaderboard badges (or use the “Copy PwC badge” button in the Results section):
Kind regards,
Niels
Hi,
Niels here from the open-source team at Hugging Face. Congratulations on your work!
I've made the paper and 6 paper-native evaluations available on Papers with Code.
The paper has results on Image Understanding and Video Understanding task pages.
The GIT2 result currently ranks first on COCO Captions Image-to-Text Retrieval.
The GIT2 results currently rank fourth on MSVD-QA, NoCaps, and TextCaps.
Would it be possible to verify these results and let me know if any score, model name, benchmark protocol, or openness metadata should be corrected?
You can also edit the task, methods, project page, and GitHub URL directly from the paper page using your Hugging Face account.
If you'd like to showcase the results in your repository README, you can copy these live leaderboard badges (or use the “Copy PwC badge” button in the Results section):
Kind regards,
Niels