Add pipeline tag, library name and link to paper (#1)
Browse files- Add pipeline tag, library name and link to paper (c3e050339ada2f027124c01a4ded23f725881d42)
Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -1,3 +1,23 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
--
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
pipeline_tag: text-to-image
|
| 4 |
+
library_name: transformers
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
# LANTERN
|
| 8 |
+
|
| 9 |
+
This repository is an official PyTorch implementation of the paper [LANTERN: Accelerating Visual Autoregressive Models via Relaxed Speculative Decoding](https://arxiv.org/abs/2410.03355) (ICLR 2025) and [LANTERN++: Enhanced Relaxed Speculative Decoding with Static Tree Drafting for Visual Auto-regressive Models](https://arxiv.org/abs/2502.06352) (ICLRW - SCOPE(Oral) 2025), which supports various functionalities related to LANTERN, including model inference, drafter model training, drafter model training data generation and image decoding for image generation.
|
| 10 |
+
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
## 📚 Table of Contents
|
| 14 |
+
1. [Directory Structure](#directory-structure)
|
| 15 |
+
2. [Installation](#installation)
|
| 16 |
+
3. [Key Features](#key-features)
|
| 17 |
+
4. [Usage](#usage)
|
| 18 |
+
5. [License](#license)
|
| 19 |
+
6. [Acknowledgement](#acknowledgement)
|
| 20 |
+
7. [Citation](#citation)
|
| 21 |
+
---
|
| 22 |
+
**(Content below remains unchanged)**
|
| 23 |
+
Code: https://github.com/jadohu/LANTERN
|