implementation detail · not yet filed
low-resolution contrastive pre-training of vision encoder
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
we restrict the maximum input resolution to 224× 224 pixels by downscaling larger images while preserving their aspect ratios.
usedunclearin DeepSeek-V4.1-FlashDeepSeek