Hi,
Niels here from the open-source team at Hugging Face. Congratulations on your work!
I've made the paper and 5 verified paper-native evaluations available on Papers with Code.
The paper has results on Document Understanding, OCR, and Image Understanding task pages.
The OV-Encoder-Lang + Qwen3-4B-Instruct2507 results currently rank second on InfoVQA and OCRBench v2.
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? The imported rows are tied to the paper or its official release artifacts; comparison-table baselines were not added.
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 5 verified paper-native evaluations available on Papers with Code.
The paper has results on Document Understanding, OCR, and Image Understanding task pages.
The OV-Encoder-Lang + Qwen3-4B-Instruct2507 results currently rank second on InfoVQA and OCRBench v2.
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? The imported rows are tied to the paper or its official release artifacts; comparison-table baselines were not added.
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