Effective pruning of web-scale datasets based on complexity of concept clusters
University of Tübingen（蒂宾根大学）；Meta AI (FAIR)（Meta AI（FAIR））；ELLIS Institute Tübingen；Max-Planck Institute for Intelligent Systems Tübingen（蒂宾根马克斯·普朗克智能系统研究所）；University of California San Diego（加州大学圣迭戈分校）；DatologyAI
Amro Abbas, Evgenia Rusak, Kushal Tirumala, Wieland Brendel, Kamalika Chaudhuri, Ari S. Morcos
Utilizing massive web-scale datasets has led to unprecedented performance gains in machine learning models, but also imposes outlandish compute requirements for their training. In order to improve training and data efficiency, we here push the limits of pruning large-scale multimodal datasets for training CLIP-style models. Today's most effective pruning method on ImageNet clusters data samples into separate concepts according to their embedding and prunes away the most prototypical samples. We scale this approach to LAION and improve it by noting that the pruning rate should be concept-specific and adapted to the complexity of the concept. Using a simple and intuitive complexity measure, we are able to reduce the training cost to a quarter of regular training. By filtering from the LAION dataset, we find that training on a smaller set of high-quality data can lead to higher performance with significantly lower training costs. More specifically, we are able to outperform the LAION-trained OpenCLIP-ViT-B32 model on ImageNet zero-shot accuracy by 1.1p.p. while only using 27.7% of the data and training compute. Despite a strong reduction in training cost, we also see improvements on ImageNet dist. shifts, retrieval tasks and VTAB. On the DataComp Medium benchmark, we achieve a new state-of-the-art Imagehttps://info.arxiv.org/help/prep#commentsNet zero-shot accuracy and a competitive average zero-shot accuracy on 38 evaluation tasks.
Structured Pruning for Deep Convolutional Neural Networks: A survey
Centre for Frontier AI Research (CFAR), Agency for Science Technology and Research (A*STAR)（新加坡科技研究局前沿人工智能研究中心）；Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局高性能计算研究所）
MADTP: Multimodal Alignment-Guided Dynamic Token Pruning for Accelerating Vision-Language Transformer
School of Information Science and Technology, Fudan University（复旦大学信息科学与技术学院）；Academy for Engineering and Technology, Fudan University（复旦大学工程技术研究院）；Tsinghua-Berkeley Shenzhen Institute, Tsinghua University（清华大学深圳国际研究生院（清华-伯克利深圳学院））
Electrostatic Force Regularization for Neural Structured Pruning
Université Polytechnique Hauts-de-France（上法兰西理工大学）；Université de Lille（里尔大学）；Université Paris Est Créteil（巴黎东克雷泰伊大学）；National Higher School of Biotechnology（国家高等生物技术学校）
Selecting Relevant Web Trained Concepts for Automated Event Retrieval
Center For Automation Research, University of Maryland, College Park（马里兰大学帕克分校自动化研究中心）
复杂事件检索是一个具有挑战性的研究问题，特别是在没有训练视频的情况下。收集训练视频的替代方案是先验地训练大型语义概念库。给定事件的文本描述，事件检索通过选择与事件描述语言相关的概念并融合看不见的视频上的概念响应来执行。然而，定义一个详尽的概念词典和预训练它需要大量的计算资源。因此，最近的方法自动化的概念发现和训练，利用大量的弱注释的Web数据。紧凑的视觉上突出的概念是自动获得的概念对，或更一般地说，n-gram的使用。然而，并不是所有视觉上突出的n-gram都一定对事件查询有用--一些概念的组合可能在视觉上紧凑但不相关--这会极大地影响性能。我们提出了一个事件检索算法，构造对自动发现的概念，然后修剪这些概念是不太可能有助于检索。修剪取决于查询和正在评估的特定视频实例。我们的方法还解决了校准和领域自适应问题时出现的概念检测器看不见的视频。我们证明了在TRECVID MED 13数据集上比其他基于视觉的系统有很大的改进。
Pruning for Robust Concept Erasing in Diffusion Models
School of Computer Science, University of Sydney（悉尼大学计算机科学学院）；Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）
Roots Beneath the Cut: Uncovering the Risk of Concept Revival in Pruning-Based Unlearning for Diffusion Models
University of Georgia（佐治亚大学）；Carnegie Mellon University（卡内基梅隆大学）；Northeastern University（东北大学）；Stevens Institute of Technology（史蒂文斯理工学院）；University of Arizona（亚利桑那大学）
