Verify evals on Papers with Code

Hi,

Niels here from the open-source team at Hugging Face. Congratulations on your work!

I've made the paper and 26 verified paper-native evaluations available on Papers with Code.

The paper has results on Image Understanding, Computer Use Agents, OCR, and 8 additional task pages.

The Nemotron 3 Nano Omni (Reasoning) result currently ranks second on OCRBench v2.

The Nemotron 3 Nano Omni result currently ranks third on ScreenSpot.

The Nemotron 3 Nano Omni (Reasoning) result currently ranks third on RefCOCO.

The Nemotron 3 Nano Omni (Reasoning) result currently ranks fourth on CharXiv (DQ).

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

Start by reviewing the linked Papers with Code paper, its 26 verified evaluations, and the listed OCRBench v2, ScreenSpot, RefCOCO, and CharXiv results. Compare the scores, model names, benchmark protocols, and openness metadata with the paper or official release artifacts. Done means reporting any discrepancies and confirming which imported results are accurate.