Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling
Paper
•
2501.16975
•
Published
•
32
Huang Hongzhi
xyzed
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
6 months ago
Frac-Connections: Fractional Extension of Hyper-Connections
authored
a paper
7 months ago
Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling
Organizations
None yet
authored
a
paper
7 months ago
authored
a
paper
10 months ago