search

yangyangyang127 / APE

[ICCV 2023] Code for "Not All Features Matter: Enhancing Few-shot CLIP with Adaptive Prior Refinement"
135 stars 12 forks source link

A Question on Formula 11 in Papers #9

Open wzhiyuan2016 opened 10 months ago

wzhiyuan2016 commented 10 months ago

hi 企业微信截图_16982910309022 The smaller the kl divergence value, the more similar it is. So it should be to increase contribution, so add a negative sign before gamma. Am I right in understanding this way?