Still There, No Longer Seen: Exposing Compression-Induced Risk in Large Vision-Language Models
CIRA induces LVLM failures only after visual-token compression, with a 20.35% mean failure rate.
1 stories, newest first · one card per story
CIRA induces LVLM failures only after visual-token compression, with a 20.35% mean failure rate.
Today 2.1K readers · 10K views · 8K reads · all time 157K views