PulseAugur
中
实时 13:29:17

Survey details adversarial attacks targeting Vision Transformer efficiency

作者 PulseAugur 编辑部 · [2 个来源] ·

一篇新的调查论文研究了通过利用 Vision Transformer（ViT）的输入自适应推理机制来降低其效率的对抗性攻击。这些攻击旨在增加计算负载，同时不显著影响准确性。该论文在 A-ViT、ATS 和 AdaViT 等各种令牌剪枝框架上比较了两种此类攻击：SlowFormer 和 DeSparsify，使用了 GFLOPs、准确率损失和攻击成功率等指标。理解这些漏洞对于开发轻量级对策以在资源受限环境中部署至关重要。 AI

影响 强调了高效 AI 模型推理中的漏洞，需要强大的防御措施以确保安全部署。

排序理由 该集群包含一篇关于 Vision Transformer 对抗性攻击研究的调查论文。

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 →

Survey details adversarial attacks targeting Vision Transformer efficiency

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
该集群包含一篇关于 Vision Transformer 对抗性攻击研究的调查论文。
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, safety
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
64 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. Hugging Face Daily Papers TIER_1 English(EN) ·

    一种利用输入自适应优化来降低Vision Transformer对抗效率退化的调查研究

    Vision Transformers (ViTs) increasingly rely on input-adaptive inference, such as token pruning and early halting, to meet energy and latency budgets. This survey examines a recent class of adversarial efficiency degradation attacks that target these mechanisms to increase comput…

  2. arXiv cs.CV TIER_1 English(EN) · Anadi Goyal, Nandish Chattopadhyay, Anupam Chattopadhyay, Chandan Karfa ·

    一种利用输入自适应优化来降低Vision Transformer对抗效率退化的调查研究

    arXiv:2608.05217v1 Announce Type: cross Abstract: Vision Transformers (ViTs) increasingly rely on input-adaptive inference, such as token pruning and early halting, to meet energy and latency budgets. This survey examines a recent class of adversarial efficiency degradation attac…

相关实体

相关话题