In the realm of neural architecture design, achieving high performance is largely reliant on the manual expertise of researchers. Despite the emergence of Neural Architecture Search (NAS) as a promising technique for automating this process, current NAS methods still require human input to expand the search space and cannot generate new architectures. This paper explores the potential of Transformers in comprehending neural architectures and their performance, with the objective of establishing the foundation for utilizing Transformers to generate novel networks. We propose the Token-based Architecture Transformer (TART), which predicts neural network performance without the need to train candidate networks. TART attains state-of-the-art performance on the DeepNets-1M dataset for performance prediction tasks without edge information, indicating the potential of Transformers to aid in discovering novel and high-performing neural architectures.
Rethinking Tokenized Graph Transformers for Node Classification
School of Computer Science and Technology, Huazhong University of Science and Technology（华中科技大学计算机科学与技术学院）；Hopcroft Center on Computing Science, Huazhong University of Science and Technology（华中科技大学霍普克罗夫特计算科学中心）；Cornell University（康奈尔大学）
KAIST（韩国科学技术院）；LG AI Research（LG人工智能研究院）；University of Illinois Chicago（伊利诺伊大学芝加哥分校）
We show that standard Transformers without graph-specific modifications can lead to promising results in graph learning both in theory and practice. Given a graph, we simply treat all nodes and edges as independent tokens, augment them with token embeddings, and feed them to a Transformer. With an appropriate choice of token embeddings, we prove that this approach is theoretically at least as expressive as an invariant graph network (2-IGN) composed of equivariant linear layers, which is already more expressive than all message-passing Graph Neural Networks (GNN). When trained on a large-scale…
Huazhong University of Science and Technology（华中科技大学）
The graph Transformer emerges as a new architecture and has shown superior performance on various graph mining tasks. In this work, we observe that existing graph Transformers treat nodes as independent tokens and construct a single long sequence composed of all node tokens so as to train the Transformer model, causing it hard to scale to large graphs due to the quadratic complexity on the number of nodes for the self-attention computation. To this end, we propose a Neighborhood Aggregation Graph Transformer (NAGphormer) that treats each node as a sequence containing a series of tokens constru…
Vision transformers rely on a patch token based self attention mechanism, in contrast to convolutional networks. We investigate fundamental differences between these two families of models, by designing a block sparsity based adversarial token attack. We probe and analyze transformer as well as convolutional models with token attacks of varying patch sizes. We infer that transformer models are more sensitive to token attacks than convolutional models, with ResNets outperforming Transformer models by up to $\sim30\%$ in robust accuracy for single token attacks.
Token Caching for Diffusion Transformer Acceleration
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences（多模态人工智能系统国家重点实验室，自动化研究所）；School of Information and Electronic Engineering, Shandong Technology and Business University（山东科技职业大学信息与电子工程学院）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；PeopleAl, Inc.（PeopleAl公司）；School of Information Science and Technology, ShanghaiTech University（上海科技大学信息科学与技术学院）；Terminal Technology Department, Alipay, Ant Group（蚂蚁集团支付宝终端技术部）
Computer vision has achieved remarkable success by (a) representing images as uniformly-arranged pixel arrays and (b) convolving highly-localized features. However, convolutions treat all image pixels equally regardless of importance; explicitly model all concepts across all images, regardless of content; and struggle to relate spatially-distant concepts. In this work, we challenge this paradigm by (a) representing images as semantic visual tokens and (b) running transformers to densely model token relationships. Critically, our Visual Transformer operates in a semantic token space, judiciousl…
Convolutional neural networks (CNN) have demonstrated outstanding Compressed Sensing (CS) performance compared to traditional, hand-crafted methods. However, they are broadly limited in terms of generalisability, inductive bias and difficulty to model long distance relationships. Transformer neural networks (TNN) overcome such issues by implementing an attention mechanism designed to capture dependencies between inputs. However, high-resolution tasks typically require vision Transformers (ViT) to decompose an image into patch-based tokens, limiting inputs to inherently local contexts. We propo…
University of Waterloo（滑铁卢大学）；Peking University（北京大学）
Transformers are powerful for sequence modeling. Nearly all state-of-the-art language models and pre-trained language models are based on the Transformer architecture. However, it distinguishes sequential tokens only with the token position index. We hypothesize that better contextual representations can be generated from the Transformer with richer positional information. To verify this, we propose a segment-aware Transformer (Segatron), by replacing the original token position encoding with a combined position encoding of paragraph, sentence, and token. We first introduce the segment-aware m…
Bridging the Dimensional Chasm: Uncover Layer-wise Dimensional Reduction in Transformers through Token Correlation
School of Physics, Peking University（北京大学物理学院）；Institute of Theoretical Physics, Chinese Academy of Sciences（中国科学院理论物理研究所）；Center for High Energy Physics, Peking University（北京大学高能物理研究中心）；Beijing Computational Science Research Center（北京计算科学研究中心）
Tsinghua University（清华大学）；UCLA（加州大学洛杉矶分校）；University of Washington（华盛顿大学）
Attention is sparse in vision transformers. We observe the final prediction in vision transformers is only based on a subset of most informative tokens, which is sufficient for accurate image recognition. Based on this observation, we propose a dynamic token sparsification framework to prune redundant tokens progressively and dynamically based on the input. Specifically, we devise a lightweight prediction module to estimate the importance score of each token given the current features. The module is added to different layers to prune redundant tokens hierarchically. To optimize the prediction …
Dalian University of Technology（大连理工大学）；Tianjin University（天津大学）；MEGVII Technology（旷视科技）
Transformer models are not only successful in natural language processing (NLP) but also demonstrate high potential in computer vision (CV). Despite great advance, most of works only focus on improvement of architectures but pay little attention to the classification head. For years transformer models base exclusively on classification token to construct the final classifier, without explicitly harnessing high-level word tokens. In this paper, we propose a novel transformer model called second-order transformer (SoT), exploiting simultaneously the classification token and word tokens for the c…
Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）；Tsinghua University（清华大学）；PhiGent Robotics（鉴智机器人）
Data mixing strategies (e.g., CutMix) have shown the ability to greatly improve the performance of convolutional neural networks (CNNs). They mix two images as inputs for training and assign them with a mixed label with the same ratio. While they are shown effective for vision transformers (ViTs), we identify a token fluctuation phenomenon that has suppressed the potential of data mixing strategies. We empirically observe that the contributions of input tokens fluctuate as forward propagating, which might induce a different mixing ratio in the output tokens. The training target computed by the…
Hourglass Tokenizer for Efficient Transformer-Based 3D Human Pose Estimation
National Key Laboratory of General Artificial Intelligence, Peking University, Shenzhen Graduate School（北京大学深圳研究生院通用人工智能国家重点实验室）；Amazon Prime Video（亚马逊Prime视频）；University of Trento（特伦托大学）
NTFormer: A Composite Node Tokenized Graph Transformer for Node Classification
School of Computer Science and Technology, Huazhong University of Science and Technology（华中科技大学计算机科学与技术学院）；Institute of Artificial Intelligence, Huazhong University of Science and Technology（华中科技大学人工智能研究院）；Hopcroft Center on Computing Science, Huazhong University of Science and Technology（华中科技大学霍普克罗夫特计算科学中心）
H$_{2}$OT: Hierarchical Hourglass Tokenizer for Efficient Video Pose Transformers
State Key Laboratory of General Artificial Intelligence, Peking University, Shenzhen Graduate School（国家一般人工智能实验室，北京大学深圳研究生院）；College of Computing and Data Science, Nanyang Technological University（计算与数据科学学院，南洋理工大学）；Amazon AGI（亚马逊人工智能实验室）；University of Trento（特伦托大学）
In the Transformer model, "self-attention" combines information from attended embeddings into the representation of the focal embedding in the next layer. Thus, across layers of the Transformer, information originating from different tokens gets increasingly mixed. This makes attention weights unreliable as explanations probes. In this paper, we consider the problem of quantifying this flow of information through self-attention. We propose two methods for approximating the attention to input tokens given attention weights, attention rollout and attention flow, as post hoc methods when we use a…
Institute of Artificial Intelligence, Huazhong University of Science & Technology（华中科技大学人工智能研究院）；School of EIC, Huazhong University of Science & Technology（华中科技大学电子信息与通信学院）；Huawei Inc.（华为公司）
Transformers have offered a new methodology of designing neural networks for visual recognition. Compared to convolutional networks, Transformers enjoy the ability of referring to global features at each stage, yet the attention module brings higher computational overhead that obstructs the application of Transformers to process high-resolution visual data. This paper aims to alleviate the conflict between efficiency and flexibility, for which we propose a specialized token for each region that serves as a messenger (MSG). Hence, by manipulating these MSG tokens, one can flexibly exchange visu…
Uncertainty-Aware Token Importance Estimation in Spiking Transformers
School of Computer Science, Peking University（北京大学计算机科学学院）；School of Computer Science, Peking University. Institute for Artificial Intelligence, Peking University（北京大学计算机科学学院。人工智能研究所）；Peking University（北京大学）
