SparseVAR is a training-free acceleration method for next-scale prediction models that excludes low-frequency tokens in high-resolution stages, yielding up to 2x speedup on Infinity-2B with a GenEval drop of 0.0051.
An image is worth 1/2 tokens after layer 2: Plug-and-play inference acceleration for large vision-language models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Frequency-Aware Autoregressive Modeling for Efficient High-Resolution Image Synthesis
SparseVAR is a training-free acceleration method for next-scale prediction models that excludes low-frequency tokens in high-resolution stages, yielding up to 2x speedup on Infinity-2B with a GenEval drop of 0.0051.