PulseAugur
中
实时 07:56:05
English(EN) One Block, Multiple Depths: Recurrent Vision Transformers with Depth-Programmed Experts

循环视觉 Transformer 用单个块实现全深度编码器精度

作者 PulseAugur 编辑部 · [2 个来源] ·

研究人员开发了一种新颖的循环视觉 Transformer (reViT)，它利用单个 Transformer 块进行递归应用，从而达到与全深度视觉编码器相当的精度。这种方法将每个递归深度的前馈网络 (FFN) 表示为共享专家的组合，这些专家由归一化的深度坐标进行编程。reViT 模型在 ImageNet-1k 上的监督训练和从 DINOv2 教师蒸馏方面均表现出色，其性能优于其他混合专家 (MoE) 系列，并以显著更少的参数实现了 DeiT III 的精度。 AI

影响 这项研究通过减少参数数量同时保持高精度，可能带来更高效的视觉模型。

排序理由 该集群描述了一篇介绍新型模型架构的最新研究论文。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

循环视觉 Transformer 用单个块实现全深度编码器精度

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
Research
该集群描述了一篇介绍新型模型架构的最新研究论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, model release
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
6 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 [2]

  1. arXiv cs.LG TIER_1 English(EN) · Adrian Bulat, Yassine Ouali, Georgios Tzimiropoulos ·

    一个块，多个深度：具有深度编程专家的循环视觉Transformer

    arXiv:2610.12448v1 Announce Type: cross Abstract: In this work, we show that a single Transformer block, applied recurrently, can match the accuracy of a full-depth vision encoder at comparable inference FLOPs without intermediate feature distillation. reViT restores depth-specif…

  2. Hugging Face Daily Papers TIER_1 English(EN) ·

    一个块，多个深度：具有深度编程专家的循环视觉Transformer

    In this work, we show that a single Transformer block, applied recurrently, can match the accuracy of a full-depth vision encoder at comparable inference FLOPs without intermediate feature distillation. reViT restores depth-specific transformations by representing the FFN at each…

相关实体

相关话题