What Generation Knows: Unlocking Fine-Grained Understanding in Unified Multimodal Models

/ Paper details

Research fromHKU

Published:: 2026.10.13· 1 min read
Authors:: Tengju RuWeitong LianZecong Tang
+1 more
Zhiheng Liu
Topics:: Other ML

Read the PDF arXiv Markdown

BibTeX

@misc{arxiv2610_23987,
  title = {What Generation Knows: Unlocking Fine-Grained Understanding in Unified Multimodal Models},
  author = {Tengju Ru and Weitong Lian and Zecong Tang and Zhiheng Liu},
  year = {2026},
  eprint = {2610.23987},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.23987}
}

/ Paper arXiv 2610.23987

/ Your notes

Select text in the abstract to highlight it or add a note. They stay in this browser.

/ Next pickEditing Images into Reusable Layers via Compositional Edit-Prior Distillation

In short

Subscribe to the free newsletter