Diffusion transformers typically incorporate textual information via attention layers and a modulation mechanism using a pooled text embedding. Nevertheless, recent approaches discard modulation-based text conditioning and rely exclusively on attention. In this paper, we address whether modulation-based text conditioning is necessary and whether it can provide any performance advantage. Our analysis shows that, in its conventional usage, the pooled embedding contributes little to overall performance, suggesting that attention alone is generally sufficient for faithfully propagating prompt information. However, we reveal that the pooled embedding can provide significant gains when used from a different perspective-serving as guidance and enabling controllable shifts toward more desirable properties. This approach is training-free, simple to implement, incurs negligible runtime overhead, and can be applied to various diffusion models, bringing improvements across diverse tasks, including text-to-image/video generation and image editing.
Not All Actions Are Equal: Rethinking Conditioning for Dexterous World Model
The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Shenzhen University（深圳大学）；Beijing University of Technology（北京工业大学）；JD Explore Academy（京东探索研究院）
Rethinking Cross-Modal Interaction in Multimodal Diffusion Transformers
The University of Hong Kong（香港大学）；Nanjing University（南京大学）；University of Chinese Academy of Sciences（中国科学院大学）；Nanyang Technological University（南洋理工大学）；Harbin Institute of Technology（哈尔滨工业大学）
National University of Singapore（新加坡国立大学）；Tencent PCG（腾讯平台与内容事业群）
Vector-Quantized (VQ-based) generative models usually consist of two basic components, i.e., VQ tokenizers and generative transformers. Prior research focuses on improving the reconstruction fidelity of VQ tokenizers but rarely examines how the improvement in reconstruction affects the generation ability of generative transformers. In this paper, we surprisingly find that improving the reconstruction fidelity of VQ tokenizers does not necessarily improve the generation. Instead, learning to compress semantic features within VQ tokenizers significantly improves generative transformers' ability …
Image-to-Image Translation with Diffusion Transformers and CLIP-Based Image Conditioning
Department of Mechanical and Aerospace Engineering（机械与航空航天工程系）；University of Houston（休斯顿大学）；School of Engineering（工程学院）；Santa Clara University（圣克拉拉大学）；School of Electrical and Computer Engineering（电气与计算机工程学院）；Cornell University（康奈尔大学）；Department of Electrical and Computer Engineering（电气与计算机工程系）；Northeastern University（东北大学）
DiT-JSCC: Rethinking Deep JSCC with Diffusion Transformers and Semantic Representations
Beijing University of Posts and Telecommunications（北京邮电大学）；Shanghai Jiao Tong University（上海交通大学）；University of Shanghai for Science and Technology（上海科技大学）
Generating visual layouts is an essential ingredient of graphic design. The ability to condition layout generation on a partial subset of component attributes is critical to real-world applications that involve user interaction. Recently, diffusion models have demonstrated high-quality generative performances in various domains. However, it is unclear how to apply diffusion models to the natural representation of layouts which consists of a mix of discrete (class) and continuous (location, size) attributes. To address the conditioning layout generation problem, we introduce DLT, a joint discre…
Enhancing Text-to-Image Diffusion Transformer via Split-Text Conditioning
Tongji University（同济大学）；The Ohio State University（俄亥俄州立大学）；Georgia Institute of Technology（佐治亚理工学院）；Shandong Academy of Sciences（山东省科学院）；Macquarie University（麦考瑞大学）
Shanghai Jiao Tong University（上海交通大学）；South China University of Technology（南方科技大学）；Beihang University（北航）；Huawei Technologies Ltd（华为技术有限公司）；Shanghai AI Lab（上海人工智能实验室）；Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））
DiC: Rethinking Conv3x3 Designs in Diffusion Models
State Key Lab of General AI, School of Intelligence Science and Technology, Peking University（通用人工智能国家重点实验室，智能科学与技术学院，北京大学）；School of Artificial Intelligence, Beijing University of Posts and Telecommunications（人工智能学院，北京邮电大学）；Huawei Noah’s Ark Lab（华为诺亚实验室）；School of Mathematical Sciences, Peking University（数学科学学院，北京大学）
Very recently, Window-based Transformers, which computed self-attention within non-overlapping local windows, demonstrated promising results on image classification, semantic segmentation, and object detection. However, less study has been devoted to the cross-window connection which is the key element to improve the representation ability. In this work, we revisit the spatial shuffle as an efficient way to build connections among windows. As a result, we propose a new vision transformer, named Shuffle Transformer, which is highly efficient and easy to implement by modifying two lines of code.…
Chongqing University of Technology（重庆理工大学）；University of California（加利福尼亚大学）
Recent advances of Transformers have brought new trust to computer vision tasks. However, on small dataset, Transformers is hard to train and has lower performance than convolutional neural networks. We make vision transformers as data-efficient as convolutional neural networks by introducing multi-focal attention bias. Inspired by the attention distance in a well-trained ViT, we constrain the self-attention of ViT to have multi-scale localized receptive field. The size of receptive field is adaptable during training so that optimal configuration can be learned. We provide empirical evidence t…
Vision Transformer (ViT) extends the application range of transformers from language processing to computer vision tasks as being an alternative architecture against the existing convolutional neural networks (CNN). Since the transformer-based architecture has been innovative for computer vision modeling, the design convention towards an effective architecture has been less studied yet. From the successful design principles of CNN, we investigate the role of spatial dimension conversion and its effectiveness on transformer-based architecture. We particularly attend to the dimension reduction p…
Technical University of Munich（慕尼黑工业大学）；East China Normal University（华东师范大学）
With the development of the self-attention mechanism, the Transformer model has demonstrated its outstanding performance in the computer vision domain. However, the massive computation brought from the full attention mechanism became a heavy burden for memory consumption. Sequentially, the limitation of memory reduces the possibility of improving the Transformer model. To remedy this problem, we propose a novel memory economy attention mechanism named Couplformer, which decouples the attention map into two sub-matrices and generates the alignment scores from spatial information. A series of di…
Vision Transformers have achieved state-of-the-art performance in many visual tasks. Due to the quadratic computational and memory complexities of self-attention, recent works either apply attention only to low-resolution inputs or restrict the receptive field to a small local region. To overcome these limitations, we propose key-only attention, which excludes query-key pairwise interactions and uses a compute-efficient saliency-gate to obtain attention weights, modeling local-global interactions in all stages. Key-only attention has linear computational and memory complexities w.r.t input siz…
Exploring sample relationships within each mini-batch has shown great potential for learning image representations. Existing works generally adopt the regular Transformer to model the visual content relationships, ignoring the cues of semantic/label correlations between samples. Also, they generally adopt the "full" self-attention mechanism which are obviously redundant and also sensitive to the noisy samples. To overcome these issues, in this paper, we design a simple yet flexible Batch-Graph Transformer (BGFormer) for mini-batch sample representations by deeply capturing the relationships of…
Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）；Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（中国科学院深圳先进技术研究院）；Shanghai AI Laboratory（上海人工智能实验室）；The University of Sydney（悉尼大学）；University of Chinese Academy of Sciences（中国科学院大学）；Tencent PCG（腾讯平台与内容事业群）
The alignment of adjacent frames is considered an essential operation in video super-resolution (VSR). Advanced VSR models, including the latest VSR Transformers, are generally equipped with well-designed alignment modules. However, the progress of the self-attention mechanism may violate this common sense. In this paper, we rethink the role of alignment in VSR Transformers and make several counter-intuitive observations. Our experiments show that: (i) VSR Transformers can directly utilize multi-frame information from unaligned videos, and (ii) existing alignment methods are sometimes harmful …
Sun Yat-sen University（中山大学）；Microsoft Research Asia（微软亚洲研究院）
Relative position encoding (RPE) is important for transformer to capture sequence ordering of input tokens. General efficacy has been proven in natural language processing. However, in computer vision, its efficacy is not well studied and even remains controversial, e.g., whether relative position encoding can work equally well as absolute position? In order to clarify this, we first review existing relative position encoding methods and analyze their pros and cons when applied in vision transformers. We then propose new relative position encoding methods dedicated to 2D images, called image R…
Rethinking Local Perception in Lightweight Vision Transformer
Tsinghua University（清华大学）；Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）
Fudan University（复旦大学）；University of Oxford（牛津大学）；University of Surrey（萨里大学）；Tencent Youtu Lab（腾讯优图实验室）；Facebook AI（脸书人工智能）
Most recent semantic segmentation methods adopt a fully-convolutional network (FCN) with an encoder-decoder architecture. The encoder progressively reduces the spatial resolution and learns more abstract/semantic visual concepts with larger receptive fields. Since context modeling is critical for segmentation, the latest efforts have been focused on increasing the receptive field, through either dilated/atrous convolutions or inserting attention modules. However, the encoder-decoder based FCN architecture remains unchanged. In this paper, we aim to provide an alternative perspective by treatin…
JD Explore Academy（京东探索研究院）；The University of Sydney（悉尼大学）
Self-supervised pre-training vision transformer (ViT) via masked image modeling (MIM) has been proven very effective. However, customized algorithms should be carefully designed for the hierarchical ViTs, e.g., GreenMIM, instead of using the vanilla and simple MAE for the plain ViT. More importantly, since these hierarchical ViTs cannot reuse the off-the-shelf pre-trained weights of the plain ViTs, the requirement of pre-training them leads to a massive amount of computational cost, thereby incurring both algorithmic and computational complexity. In this paper, we address this problem by propo…
DETR is a recently proposed Transformer-based method which views object detection as a set prediction problem and achieves state-of-the-art performance but demands extra-long training time to converge. In this paper, we investigate the causes of the optimization difficulty in the training of DETR. Our examinations reveal several factors contributing to the slow convergence of DETR, primarily the issues with the Hungarian loss and the Transformer cross-attention mechanism. To overcome these issues we propose two solutions, namely, TSP-FCOS (Transformer-based Set Prediction with FCOS) and TSP-RC…
School of Computer Science and Software Engineering, Southwest Petroleum University（计算机科学与软件工程学院，西南石油大学）；School of Computing and Artificial Intelligence, Southwest Jiaotong University（计算与人工智能学院，西南交通大学）；School of Computing and Artificial Intelligence, Southwestern University of Finance and Economics（计算与人工智能学院，西南财经大学）
To ensure proper knowledge representation of the kitchen environment, it is vital for kitchen robots to recognize the states of the food items that are being cooked. Although the domain of object detection and recognition has been extensively studied, the task of object state classification has remained relatively unexplored. The high intra-class similarity of ingredients during different states of cooking makes the task even more challenging. Researchers have proposed adopting Deep Learning based strategies in recent times, however, they are yet to achieve high performance. In this study, we …
The Chinese University of Hong Kong（香港中文大学）；National University of Singapore（新加坡国立大学）；Imperial College London（帝国理工学院）；National University of Singapore (Suzhou) Research Institute（新加坡国立大学（苏州）研究院）
Surgical captioning plays an important role in surgical instruction prediction and report generation. However, the majority of captioning models still rely on the heavy computational object detector or feature extractor to extract regional features. In addition, the detection model requires additional bounding box annotation which is costly and needs skilled annotators. These lead to inference delay and limit the captioning model to deploy in real-time robotic surgery. For this purpose, we design an end-to-end detector and feature extractor-free captioning model by utilizing the patch-based sh…
UltraImage: Rethinking Resolution Extrapolation in Image Diffusion Transformers
Dept. of Comp. Sci. & Tech., BNRist Center, Tsinghua University（计算机科学与技术系，北京理工大学中心，清华大学）；ShengShu（盛书）；Gaoling School of Artificial Intelligence, Renmin University of China（光明人工智能学院，中国人民大学）；Princeton University（普林斯顿大学）
Huazhong University of Science & Technology（华中科技大学）；Horizon Robotics（地平线机器人）
Can Transformer perform 2D object- and region-level recognition from a pure sequence-to-sequence perspective with minimal knowledge about the 2D spatial structure? To answer this question, we present You Only Look at One Sequence (YOLOS), a series of object detection models based on the vanilla Vision Transformer with the fewest possible modifications, region priors, as well as inductive biases of the target task. We find that YOLOS pre-trained on the mid-sized ImageNet-1k dataset only can already achieve quite competitive performance on the challenging COCO object detection benchmark, e.g., Y…
Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院，清华大学）；ByteDance Inc.（字节跳动公司）
The vanilla self-attention mechanism inherently relies on pre-defined and steadfast computational dimensions. Such inflexibility restricts it from possessing context-oriented generalization that can bring more contextual cues and global representations. To mitigate this issue, we propose a Scalable Self-Attention (SSA) mechanism that leverages two scaling factors to release dimensions of query, key, and value matrices while unbinding them with the input. This scalability fetches context-oriented generalization and enhances object sensitivity, which pushes the whole network into a more effectiv…
Rethinking Attention Gated with Hybrid Dual Pyramid Transformer-CNN for Generalized Segmentation in Medical Imaging
Junia（朱尼亚学院）；Centrale Lille（里尔中央理工学院）；University of the Basque Country UPV/EHU（巴斯克大学 UPV/EHU）；Université Polytechnique Hauts-de-France（上法兰西理工大学）；Université de Lille（里尔大学）；Ho Chi Minh City Open University（胡志明市开放大学）
VPNeXt -- Rethinking Dense Decoding for Plain Vision Transformer
School of Information and Software Engineering, University of Electronic Science and Technology of China（信息与软件工程学院，电子科学与技术大学）；Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China（深圳先进研究所，电子科学与技术大学）
Great Bay University（大湾区大学）；Nanyang Technological University（南洋理工大学）；University of Oulu（奥卢大学）；Lappeenranta-Lahti University of Technology（拉彭兰塔-拉赫蒂理工大学）；South China University of Technology（华南理工大学）
Recently, vision transformer (ViT) based multimodal learning methods have been proposed to improve the robustness of face anti-spoofing (FAS) systems. However, there are still no works to explore the fundamental natures (\textit{e.g.}, modality-aware inputs, suitable multimodal pre-training, and efficient finetuning) in vanilla ViT for multimodal FAS. In this paper, we investigate three key factors (i.e., inputs, pre-training, and finetuning) in ViT for multimodal FAS with RGB, Infrared (IR), and Depth. First, in terms of the ViT inputs, we find that leveraging local feature descriptors benefi…
BCTR: Bidirectional Conditioning Transformer for Scene Graph Generation
Samsung R&D Institute China–Beijing（三星中国北京研发中心）；School of Mathematics, Southwestern University of Finance and Economics（西南财经大学数学学院）；State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所多模态人工智能系统国家重点实验室）；Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所复杂系统认知与决策智能实验室）；Beijing Academy of Artificial Intelligence（北京人工智能研究院）
Huazhong University of Science and Technology（华中科技大学）；Hong Kong University of Science and Technology（香港科技大学）
Vision transformers have recently set off a new wave in the field of medical image analysis due to their remarkable performance on various computer vision tasks. However, recent hybrid-/transformer-based approaches mainly focus on the benefits of transformers in capturing long-range dependency while ignoring the issues of their daunting computational complexity, high training costs, and redundant dependency. In this paper, we propose to employ adaptive pruning to transformers for medical image segmentation and propose a lightweight and effective hybrid network APFormer. To our best knowledge, …
