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](https://paperswithcode.co/paper/2512.13874) and [2 verified paper-native evaluations](https://paperswithcode.co/paper/2512.13874#results) available on Papers with Code.

The paper is part of the [Video Understanding](https://paperswithcode.co/tasks/video-understanding) task page.

The SAGE-Flash (SAGE-MM: Qwen3-VL-8B-Instruct [+SFT] [+RL]) result currently ranks third on [VideoMMMU](https://paperswithcode.co/benchmark/videommmu?task=video-understanding&eval=24516).

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):

[](https://paperswithcode.co/api/v1/papers/2512.13874/leaderboard-badge-link?eval=24516)

Kind regards,

Niels

Start with the linked Papers with Code paper's Results section and the VideoMMMU leaderboard entry. Verify the two paper-native rows—score, model name, benchmark protocol, and openness metadata—against the paper or official release artifacts, then report or correct any discrepancies.

由索引模型根据 Issue 内容生成。