Missing details on Aria Digital Twin dataset evaluation subset

Hello SAM-3D!

Thank you for releasing this amazing work.
I would like to ask the subset details of Aria Digital Twin evaluation reported in the paper.

In Appendix D.2, [the paper](https://arxiv.org/pdf/2511.16624) says:

> **Aria Digital Twin (Pan et al., 2023)**: We sample a smaller set of 40 video frames, with around 30 objects per scene.

Could you share the details of **the video indices and the frame selection scheme to collect 40 video frames?** This would significantly help follow-up research. Thank you so much !