Add model card
#1
by
nielsr HF Staff - opened
README.md
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pipeline_tag: feature-extraction
|
| 3 |
+
license: mit
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# CODA: Repurposing Continuous VAEs for Discrete Tokenization
|
| 7 |
+
|
| 8 |
+
This repository contains the CODA tokenizer, as introduced in [CODA: Repurposing Continuous VAEs for Discrete Tokenization](https://huggingface.co/papers/2503.17760).
|
| 9 |
+
|
| 10 |
+
Project Page: https://lzy-tony.github.io/coda
|
| 11 |
+
|
| 12 |
+
Code: https://github.com/LeapLabTHU/CODA
|
| 13 |
+
|
| 14 |
+
## Highlights
|
| 15 |
+
|
| 16 |
+
CODA addresses the challenges of training conventional VQ tokenizers by decoupling compression and discretization. Instead of training from scratch, CODA adapts off-the-shelf continuous VAEs into discrete tokenizers, leading to stable and efficient training with strong visual fidelity.
|