1 citations · 2 across the 10 of their papers we have counts for
1 paper · 1 filter
Junming Liu, Yuqi Li, Yifei Sun +6
Vision-Language Models (VLMs) have made striking progress, yet their spatial reasoning remains fragile. Models that answer an original input correctly can still fail under valid tr…