A sub-billion vision-language model using a fixed 9x reshape of visual tokens reports faster edge inference and higher benchmark scores than nanoLLAVA.
MLC-LLM, 2023-2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
OmniVLM: A Token-Compressed, Sub-Billion-Parameter Vision-Language Model for Efficient On-Device Inference
A sub-billion vision-language model using a fixed 9x reshape of visual tokens reports faster edge inference and higher benchmark scores than nanoLLAVA.