Learning 4d Reasoning

Enable VLMs to perform dynamic spatial reasoning (DSR) by extracting 4D priors from videos and using Geometry Selection Modules (GSM) for selective injection. Provides DSR-Train dataset (50K QA pairs) and benchmark with six reasoning types, balancing geometric specialization with general video understanding—improving VLM 4D reasoning without degradation on general tasks.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit 3b2dee5e62

Frequently asked questions

npx get-skill add adu2021/learning-4d-reasoning