Italiano(IT) OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers

OrbitQuant 实现扩散 Transformer 的数据无关量化

作者 PulseAugur 编辑部 · [1 个来源] · 2026-07-02 00:00

研究人员开发了 OrbitQuant，一种用于扩散 Transformer (DiTs) 的新型训练后量化方法。该技术通过在归一化、旋转基中进行量化来实现高效推理，无需在不同的时间步、提示或模态之间进行重新校准。OrbitQuant 在 FLUX.1、Z-Image-Turbo、Wan 2.1 和 CogVideoX 等图像和视频生成模型的低比特设置下实现了最先进的性能，甚至在 2 位权重下也能实现可用的生成质量。 AI

影响降低了扩散模型推理的计算成本，可能使其能够在资源受限的设备上更广泛地部署。

排序理由详细介绍模型量化新方法的学术论文。 [lever_c_demoted from research: ic=1 ai=1.0]

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍模型量化新方法的学术论文。 [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, infra; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 101 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

Hugging Face Daily Papers TIER_1 Italiano(IT) · 2026-07-02 00:00

OrbitQuant：图像和视频扩散 Transformer 的数据无关量化

OrbitQuant enables efficient post-training quantization for diffusion transformers by using a normalized rotated basis that eliminates the need for recalibration across different timesteps and modalities.

本文如何被排名

报道来源 [1]

OrbitQuant：图像和视频扩散 Transformer 的数据无关量化

相关实体

相关话题