Reproduce “VidUEU-Agent: A Data-Curation Agent for Multimodal Understanding, Editing, and Unified Tasks” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Tengju Ru, Weitong Lian, Zecong Tang, Lingyi Meng, Haoran Li, Zhejun Cui, Yichen Zhu, Hangshuo Cao and 7 more

VidUEU-Agent converts video into traceable instruction examples for understanding, editing, and unified tasks. It filters shots and frames, selects keyframes using tool evidence, and structures metadata to guide instruction generation; training feedback adjusts example distributions. Fine-tuning Bagel on its unified subset raises WEAVEbench Overall from 0.446 to 0.501, supporting a gain on that evaluation, while long interleaved chains remain challenging.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.