paper-with-me

Papers

Accelerating Video Generation Inference with Sequential-Parallel 3D Positional Encoding Using a Global Time Index

2026-03-02 · Chao Yuan, Pan Li arxiv

Diffusion Transformer (DiT)-based video generation models inherently suffer from bottlenecks in long video synthesis and real-time inference, which can be attributed to the use of full spatiotemporal attention. Specifically, this mechanism leads to explosive O(N^2) memory consumption and high first-frame latency. To address these issues, we implement system-level inference optimizations for a causal autoregressive video generation pipeline. We adapt the Self-Forcing causal autoregressive framework to sequence parallel inference and implement a sequence-parallel variant of the causal rotary position embedding which we refer to as Causal-RoPE SP. This adaptation enables localized computation and reduces cross-rank communication in sequence parallel execution. In addition, computation and communication pipelines are optimized through operator fusion and RoPE precomputation. Experiments conducted on an eight GPU A800 cluster show that the optimized system achieves comparable generation quality, sub-second first-frame latency, and near real-time inference speed. For generating five second 480P videos, a 1.58x speedup is achieved, thereby providing effective support for real-time interactive applications.

📄 PDF Abstract BibTeX arXiv:2603.06664

Code (0)

등록된 구현이 없습니다.

Tasks

Video Generation

Similar Papers 제목 키워드 기반

Lightning Fast Caching-based Parallel Denoising Prediction for Accelerating Talking Head Generation

2025-08-25 · Jianzhi Long, Wenhao Sun, Rongcheng Tu, Dacheng Tao arxiv

Diffusion-based talking head models generate high-quality, photorealistic videos but suffer from slow inference, limiting practical applications. Existing acceleration methods for general diffusion models fail to exploit…

Talking Head Generation

Communication-Efficient Diffusion Denoising Parallelization via Reuse-then-Predict Mechanism

2025-05-20 · Kunyun Wang, Bohan Li, Kai Yu, Minyi Guo 외

Diffusion models have emerged as a powerful class of generative models across various modalities, including image, video, and audio synthesis. However, their deployment is often limited by significant inference latency, …

Audio SynthesisDenoising

DeepPCR: Parallelizing Sequential Operations in Neural Networks

2023-09-28 · NeurIPS 2023 11

Parallelization techniques have become ubiquitous for accelerating inference and training of deep neural networks. Despite this, several operations are still performed in a sequential manner. For instance, the forward an…

Denoising

Parallel Prefix Verification for Speculative Generation

2026-05-05 · Yuncheng Yao, Yuxuan Xia, Shengjie Wang, Danyang Zhuo arxiv

We introduce PARSE (PArallel pRefix Speculative Engine), a speculative generation framework that accelerates large language model (LLM) inference by parallelizing prefix verification on a semantic level. Existing specula…

NUWA-XL: Diffusion over Diffusion for eXtremely Long Video Generation

2023-03-22 · Shengming Yin, Chenfei Wu, Huan Yang, JianFeng Wang 외

In this paper, we propose NUWA-XL, a novel Diffusion over Diffusion architecture for eXtremely Long video generation. Most current work generates long videos segment by segment sequentially, which normally leads to the g…

Video Generation