Visual representations in VLMs converge quickly to stable low-complexity forms while text continues evolving, with task-dependent needs for sustained image token access.
Pore: Position-reweighted visual token pruning for vision language models.arXiv e-prints, pages arXiv–2508
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Do Vision Language Models Need to Process Image Tokens?
Visual representations in VLMs converge quickly to stable low-complexity forms while text continues evolving, with task-dependent needs for sustained image token access.