vbercy/g2tm-segmenter

Graph-Guided Token Merging (G2TM) is a lightweight one-shot module designed to eliminate redundant tokens early in the ViT architecture. It performs a single merging step after a shallow attention block, enabling all subsequent layers to operate on a compact token set. It leverages graph theory to identify groups of semantically redundant patches.

24
/ 100
Experimental
No Package No Dependents
Maintenance 13 / 25
Adoption 2 / 25
Maturity 9 / 25
Community 0 / 25

How are scores calculated?

Stars

2

Forks

Language

Python

License

Apache-2.0

Last pushed

Mar 20, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/vbercy/g2tm-segmenter"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.