architecture · first seen 22 Oct 2026
recurrent vision encoder
A visual encoder that reuses blocks in loops to iteratively refine visual representations.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Loop to Look Better: Learning Adaptive Depth for Question-Guided Perception in Vision-Language ModelsPapers · 22 Oct 2026 · Wencheng Ye, Anning Hu, Hongye Fang, Tianyi Wang, Junchi Yan
Latest sources
- Loop to Look Better: Learning Adaptive Depth for Question-Guided Perception in Vision-Language ModelsPapers · 22 Oct 2026 · Wencheng Ye, Anning Hu, Hongye Fang, Tianyi Wang, Junchi Yan