A break-even study of Qwen2.5-VL on two GPUs finds that a post-vision pruner and a pre-vision resolution rule both cut end-to-end latency, but pre-vision routing wins by skipping vision encoding, not by removing the most tokens.
AdaptVision : Efficient vision-language models via adaptive visual acquisition
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
When Do Fewer Visual Tokens Accelerate Multimodal Inference? A Break-Even Study Across Decision Locations and Hardware
A break-even study of Qwen2.5-VL on two GPUs finds that a post-vision pruner and a pre-vision resolution rule both cut end-to-end latency, but pre-vision routing wins by skipping vision encoding, not by removing the most tokens.