FastVLM is here to show you something different! ✨



They attached MLP to FastViTHD and derived visual tokens in the LLM world.

Result? The number of tokens has significantly decreased, 4 times less than FastViT, 16 times less than ViT‑L/14, resolution 336 pixels!😲

The tokens have decreased, and the complexity has also lowered; it's simply the successful weight loss of the token world! 🤣📉
MLP1.42%
View Original
post-image
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • 6
  • Repost
  • Share
Comment
0/400
AirdropCollectorvip
· 8h ago
The dimensionality reduction attack is extremely powerful.
View OriginalReply0
UncommonNPCvip
· 8h ago
amazing compression approach
View OriginalReply0
BTCRetirementFundvip
· 8h ago
The provincial Token is truly elegant.
View OriginalReply0
BearMarketBarbervip
· 8h ago
amazing weight loss plan
View OriginalReply0
AlwaysAnonvip
· 8h ago
The large model has slimmed down again.
View OriginalReply0
AirdropBuffetvip
· 8h ago
Token weight loss is a good thing!
View OriginalReply0
Trade Crypto Anywhere Anytime
qrCode
Scan to download Gate App
Community
English
  • 简体中文
  • English
  • Tiếng Việt
  • 繁體中文
  • Español
  • Русский
  • Français (Afrique)
  • Português (Portugal)
  • Bahasa Indonesia
  • 日本語
  • بالعربية
  • Українська
  • Português (Brasil)