Pre-training vision models for the classification of alerts from wide-field time-domain surveys
Paper • 2512.11957 • Published
Pre-trained off-the-shelf BTSbots
ConvNeXt and MaxViT models with various pre-training regimens fine-tuned on the BTSbot task.
-
-
nabeelr/BTSbot-convnext-pico-randinit
Image Classification • Updated -
nabeelr/BTSbot-convnext-pico-in1k
Image Classification • Updated -
nabeelr/BTSbot-convnext-pico-galaxyzoo
Image Classification • Updated • 5 -
nabeelr/BTSbot-maxvit-tiny-galaxyzoo
Image Classification • Updated -
nabeelr/BTSbot-maxvit-tiny-in1k
Image Classification • Updated • 1 -
nabeelr/BTSbot-maxvit-tiny-randinit
Image Classification • Updated • 3 -
nabeelr/BTSbot-convnext-pico-galaxyzoo-metadata
Image Classification • Updated • 1 -
nabeelr/BTSbot-convnext-pico-in1k-metadata
Image Classification • Updated -
nabeelr/BTSbot-convnext-pico-randinit-metadata
Image Classification • Updated • 1 -
nabeelr/BTSbot-maxvit-tiny-galaxyzoo-metadata
Image Classification • Updated -
nabeelr/BTSbot-maxvit-tiny-in1k-metadata
Image Classification • Updated -
nabeelr/BTSbot-maxvit-tiny-randinit-metadata
Image Classification • Updated