IAAILoopVL
LoopVL: Recurrent Visual Intelligence
Extends looped transformers to vision-language models trained from scratch, combining Module-Loop and Model-Loop computation to iteratively update a unified vision-language state through shared modules.
Inside the method
[ Shared modules (Module-Loop) ] within [ Model-Loop ] × r over a vision-language state
Simplified conceptual schematic. Consult the paper for the complete architecture.
- Recurrence family
- Full-stack recurrence
- Depth control
- Fixed loops
- KV / state strategy
- See paper
Reading note
Trained through language pretraining, multimodal training and post-training, it outperforms several similar-size and larger non-recurrent models on multimodal understanding and visual reasoning, with visible shifts in visual attention across loops.
Sources checked 2026-10-28. This catalog does not imply independent reproduction.
Related works
Cite this work
@misc{loopvl2026,
title = {LoopVL: Recurrent Visual Intelligence},
author = {Zhe Qian and Ziyang Gong and Zhongxing Xu and Hehan Li and Zhonghua Wang and Hanmeng Liu and Liwei Qian and Xin Pei and Peizhi Xu and Fei Luo and Mingxuan Wang and Xue Yang and Shiwei liu and Yanbiao Ma and Junchi Yan and Jungong Han},
year = {2026},
eprint = {2609.38426},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.38426}
}