We propose RanDumb to examine the efficacy of continual representation learning. RanDumb embeds raw pixels using a fixed random transform which approximates an RBF-Kernel, initialized before seeing any data, and learns a simple linear classifier on top. We present a surprising and consistent finding: RanDumb significantly outperforms the continually learned representations using deep networks across numerous continual learning benchmarks, demonstrating the poor performance of representation learning in these scenarios. RanDumb stores no exemplars and performs a single pass over the data, processing one sample at a time. It complements GDumb, operating in a low-exemplar regime where GDumb has especially poor performance. We reach the same consistent conclusions when RanDumb is extended to scenarios with pretrained models replacing the random transform with pretrained feature extractor. Our investigation is both surprising and alarming as it questions our understanding of how to effectively design and train models that require efficient continual representation learning, and necessitates a principled reinvestigation of the widely explored problem formulation itself. Our code is available at //github.com/drimpossible/RanDumb.
相關內容
讓 iOS 8 和 OS X Yosemite 無縫切換的一個新特性。
> Apple products have always been designed to work together beautifully. But now they may really surprise you. With iOS 8 and OS X Yosemite, you’ll be able to do more wonderful things than ever before.
Source:
Source:
Conditional diffusion models can create unseen images in various settings, aiding image interpolation. Interpolation in latent spaces is well-studied, but interpolation with specific conditions like text or poses is less understood. Simple approaches, such as linear interpolation in the space of conditions, often result in images that lack consistency, smoothness, and fidelity. To that end, we introduce a novel training-free technique named Attention Interpolation via Diffusion (AID). Our key contributions include 1) proposing an inner/outer interpolated attention layer; 2) fusing the interpolated attention with self-attention to boost fidelity; and 3) applying beta distribution to selection to increase smoothness. We also present a variant, Prompt-guided Attention Interpolation via Diffusion (PAID), that considers interpolation as a condition-dependent generative process. This method enables the creation of new images with greater consistency, smoothness, and efficiency, and offers control over the exact path of interpolation. Our approach demonstrates effectiveness for conceptual and spatial interpolation. Code and demo are available at //github.com/QY-H00/attention-interpolation-diffusion.
Diffusion Transformer (DiT) has emerged as the new trend of generative diffusion models on image generation. In view of extremely slow convergence in typical DiT, recent breakthroughs have been driven by mask strategy that significantly improves the training efficiency of DiT with additional intra-image contextual learning. Despite this progress, mask strategy still suffers from two inherent limitations: (a) training-inference discrepancy and (b) fuzzy relations between mask reconstruction & generative diffusion process, resulting in sub-optimal training of DiT. In this work, we address these limitations by novelly unleashing the self-supervised discrimination knowledge to boost DiT training. Technically, we frame our DiT in a teacher-student manner. The teacher-student discriminative pairs are built on the diffusion noises along the same Probability Flow Ordinary Differential Equation (PF-ODE). Instead of applying mask reconstruction loss over both DiT encoder and decoder, we decouple DiT encoder and decoder to separately tackle discriminative and generative objectives. In particular, by encoding discriminative pairs with student and teacher DiT encoders, a new discriminative loss is designed to encourage the inter-image alignment in the self-supervised embedding space. After that, student samples are fed into student DiT decoder to perform the typical generative diffusion task. Extensive experiments are conducted on ImageNet dataset, and our method achieves a competitive balance between training cost and generative capacity.
Recent advancements in diffusion models have positioned them at the forefront of image generation. Despite their superior performance, diffusion models are not without drawbacks; they are characterized by complex architectures and substantial computational demands, resulting in significant latency due to their iterative sampling process. To mitigate these limitations, we introduce a dual approach involving model miniaturization and a reduction in sampling steps, aimed at significantly decreasing model latency. Our methodology leverages knowledge distillation to streamline the U-Net and image decoder architectures, and introduces an innovative one-step DM training technique that utilizes feature matching and score distillation. We present two models, SDXS-512 and SDXS-1024, achieving inference speeds of approximately 100 FPS (30x faster than SD v1.5) and 30 FP (60x faster than SDXL) on a single GPU, respectively. Moreover, our training approach offers promising applications in image-conditioned control, facilitating efficient image-to-image translation.
Multimodal Large Language Models (MLLMs) have showcased impressive skills in tasks related to visual understanding and reasoning. Yet, their widespread application faces obstacles due to the high computational demands during both the training and inference phases, restricting their use to a limited audience within the research and user communities. In this paper, we investigate the design aspects of Multimodal Small Language Models (MSLMs) and propose an efficient multimodal assistant named Mipha, which is designed to create synergy among various aspects: visual representation, language models, and optimization strategies. We show that without increasing the volume of training data, our Mipha-3B outperforms the state-of-the-art large MLLMs, especially LLaVA-1.5-13B, on multiple benchmarks. Through detailed discussion, we provide insights and guidelines for developing strong MSLMs that rival the capabilities of MLLMs. Our code is available at //github.com/zhuyiche/llava-phi.
Current training pipelines in object recognition neglect Hue Jittering when doing data augmentation as it not only brings appearance changes that are detrimental to classification, but also the implementation is inefficient in practice. In this study, we investigate the effect of hue variance in the context of video understanding and find this variance to be beneficial since static appearances are less important in videos that contain motion information. Based on this observation, we propose a data augmentation method for video understanding, named Motion Coherent Augmentation (MCA), that introduces appearance variation in videos and implicitly encourages the model to prioritize motion patterns, rather than static appearances. Concretely, we propose an operation SwapMix to efficiently modify the appearance of video samples, and introduce Variation Alignment (VA) to resolve the distribution shift caused by SwapMix, enforcing the model to learn appearance invariant representations. Comprehensive empirical evaluation across various architectures and different datasets solidly validates the effectiveness and generalization ability of MCA, and the application of VA in other augmentation methods. Code is available at //github.com/BeSpontaneous/MCA-pytorch.
Four-dimensional cone-beam computed tomography (4D CBCT) provides respiration-resolved images and can be used for image-guided radiation therapy. However, the ability to reveal respiratory motion comes at the cost of image artifacts. As raw projection data are sorted into multiple respiratory phases, there is a limited number of cone-beam projections available for image reconstruction. Consequently, the 4D CBCT images are covered by severe streak artifacts. Although several deep learning-based methods have been proposed to address this issue, most algorithms employ ordinary network models, neglecting the intrinsic structural prior within 4D CBCT images. In this paper, we first explore the origin and appearance of streak artifacts in 4D CBCT images.Specifically, we find that streak artifacts exhibit a periodic rotational motion along with the patient's respiration. This unique motion pattern inspires us to distinguish the artifacts from the desired anatomical structures in the spatiotemporal domain. Thereafter, we propose a spatiotemporal neural network named RSTAR-Net with separable and circular convolutions for Rotational Streak Artifact Reduction. The specially designed model effectively encodes dynamic image features, facilitating the recovery of 4D CBCT images. Moreover, RSTAR-Net is also lightweight and computationally efficient. Extensive experiments substantiate the effectiveness of our proposed method, and RSTAR-Net shows superior performance to comparison methods.
We present Surf-D, a novel method for generating high-quality 3D shapes as Surfaces with arbitrary topologies using Diffusion models. Previous methods explored shape generation with different representations and they suffer from limited topologies and poor geometry details. To generate high-quality surfaces of arbitrary topologies, we use the Unsigned Distance Field (UDF) as our surface representation to accommodate arbitrary topologies. Furthermore, we propose a new pipeline that employs a point-based AutoEncoder to learn a compact and continuous latent space for accurately encoding UDF and support high-resolution mesh extraction. We further show that our new pipeline significantly outperforms the prior approaches to learning the distance fields, such as the grid-based AutoEncoder, which is not scalable and incapable of learning accurate UDF. In addition, we adopt a curriculum learning strategy to efficiently embed various surfaces. With the pretrained shape latent space, we employ a latent diffusion model to acquire the distribution of various shapes. Extensive experiments are presented on using Surf-D for unconditional generation, category conditional generation, image conditional generation, and text-to-shape tasks. The experiments demonstrate the superior performance of Surf-D in shape generation across multiple modalities as conditions. Visit our project page at //yzmblog.github.io/projects/SurfD/.
The DNS HTTPS resource record is a new DNS record type designed for the delivery of configuration information and parameters required to initiate connections to HTTPS network services. It provides the ability to perform zone apex redirection to a third-party provider, which the existing CNAME record cannot do. In addition, it is a key enabler for TLS Encrypted ClientHello (ECH) by providing the cryptographic keying material needed to encrypt the initial exchange. To understand the adoption and security of this new DNS HTTPS record, we perform a longitudinal study on the server-side deployment of DNS HTTPS for Tranco top 1 million domains over 8 months, as well as the client-side support for DNS HTTPS from major browsers. To the best of knowledge, our work is the first longitudinal study on DNS HTTPS server deployment, and the first known study on client-side support for DNS HTTPS. Despite the rapidly growing trend of DNS HTTPS adoption, our study highlights concerns in the deployment by both servers and clients, such as the complexity in properly maintaining HTTPS records and the concerning hardfail mechanisms in browser when using HTTPS records.
We introduce Videoshop, a training-free video editing algorithm for localized semantic edits. Videoshop allows users to use any editing software, including Photoshop and generative inpainting, to modify the first frame; it automatically propagates those changes, with semantic, spatial, and temporally consistent motion, to the remaining frames. Unlike existing methods that enable edits only through imprecise textual instructions, Videoshop allows users to add or remove objects, semantically change objects, insert stock photos into videos, etc. with fine-grained control over locations and appearance. We achieve this through image-based video editing by inverting latents with noise extrapolation, from which we generate videos conditioned on the edited image. Videoshop produces higher quality edits against 6 baselines on 2 editing benchmarks using 10 evaluation metrics.
Researchers have proposed to use data of human preference feedback to fine-tune text-to-image generative models. However, the scalability of human feedback collection has been limited by its reliance on manual annotation. Therefore, we develop and test a method to automatically annotate user preferences from their spontaneous facial expression reaction to the generated images. We collect a dataset of Facial Expression Reaction to Generated Images (FERGI) and show that the activations of multiple facial action units (AUs) are highly correlated with user evaluations of the generated images. Specifically, AU4 (brow lowerer) is reflective of negative evaluations of the generated image whereas AU12 (lip corner puller) is reflective of positive evaluations. These can be useful in two ways. Firstly, we can automatically annotate user preferences between image pairs with substantial difference in these AU responses with an accuracy significantly outperforming state-of-the-art scoring models. Secondly, directly integrating the AU responses with the scoring models improves their consistency with human preferences. Finally, this method of automatic annotation with facial expression analysis can be potentially generalized to other generation tasks. The code is available at //github.com/ShuangquanFeng/FERGI, and the dataset is also available at the same link for research purposes.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191