Update README.md

README.md

@@ -1,4 +1,20 @@
-## HiAR: Efficient Autoregressive Long Video Generation via Hierarchical Denoising ##
+<p align="center">
+<h1 align="center">HiAR</h1>
+<h3 align="center">Hierarchical Autoregressive Video Generation with Pipelined Parallel Inference</h3>
+</p>
+<p align="center">
+  <h3 align="center"><a href="#">Paper</a> | <a href="#">Website</a> | <a href="#">Models (HuggingFace)</a></h3>
+</p>
+
+---
+
+HiAR proposes **hierarchical denoising** for autoregressive video diffusion models, a paradigm shift from conventional block-first to **step-first** denoising order. By conditioning each block on context at a matched noise level, HiAR maximally attenuates error propagation while preserving temporal causality, achieving **state-of-the-art long video generation** (20s+) with significantly reduced quality drift.
+
+Key features:
+- **Hierarchical Denoising**: Step-first denoising order with noisy context conditioning at matched noise levels
+- **Pipelined Parallel Inference**: Exploits the hierarchical structure for wall-clock speedup via multi-GPU pipeline parallelism
+- **Forward-KL Regularization**: Prevents low-motion shortcuts in reverse-KL distillation
+- **4-step generation**: Real-time streaming video generation on a single GPU
 
 ---
 license: mit