中文
相关论文

相关论文: On the Limits of Latent Reuse in Diffusion Models

200 篇论文

Standard Latent Diffusion Models rely on a complex, three-part architecture consisting of a separate encoder, decoder, and diffusion network, which are trained in multiple stages. This modular design is computationally inefficient, leads to…

计算机视觉与模式识别 · 计算机科学 2025-11-19 Xiyuan Wang , Muhan Zhang

Generative data augmentation with latent diffusion models is a promising strategy for addressing class imbalance in medical imaging, yet current approaches focus on perceptual fidelity and domain-specific autoencoder fine-tuning while…

计算机视觉与模式识别 · 计算机科学 2026-05-19 Mischa Dombrowski , Felix Nützel , Bernhard Kainz

We introduce LumiNet, a novel architecture that leverages generative models and latent intrinsic representations for effective lighting transfer. Given a source image and a target lighting image, LumiNet synthesizes a relit version of the…

计算机视觉与模式识别 · 计算机科学 2025-09-04 Xiaoyan Xing , Konrad Groh , Sezer Karaoglu , Theo Gevers , Anand Bhattad

Diffusion models, such as Stable Diffusion, have shown incredible performance on text-to-image generation. Since text-to-image generation often requires models to generate visual concepts with fine-grained details and attributes specified…

计算机视觉与模式识别 · 计算机科学 2024-04-26 Xuehai He , Weixi Feng , Tsu-Jui Fu , Varun Jampani , Arjun Akula , Pradyumna Narayana , Sugato Basu , William Yang Wang , Xin Eric Wang

Domain adaptation addresses the common problem when the target distribution generating our test data drifts from the source (training) distribution. While absent assumptions, domain adaptation is impossible, strict conditions, e.g.…

机器学习 · 计算机科学 2019-03-13 Yifan Wu , Ezra Winston , Divyansh Kaushik , Zachary Lipton

For the last decade, there has been a push to use multi-dimensional (latent) spaces to represent concepts; and yet how to manipulate these concepts or reason with them remains largely unclear. Some recent methods exploit multiple latent…

计算机视觉与模式识别 · 计算机科学 2024-09-24 Lorenzo Olearo , Giorgio Longari , Simone Melzi , Alessandro Raganato , Rafael Peñaloza

Diffusion models have gained significant attention for high-fidelity image generation. Our work investigates the potential of exploiting diffusion models for adversarial robustness in image classification and object detection. Adversarial…

图像与视频处理 · 电气工程与系统科学 2025-11-05 Mika Yagoda , Shady Abu-Hussein , Raja Giryes

Data fusion and transfer learning are rapidly growing fields that enhance model performance for a target population by leveraging other related data sources or tasks. The challenges lie in the various potential heterogeneities between the…

机器学习 · 统计学 2025-08-19 Jing Wang , HaiYing Wang , Kun Chen

Adversarial attacks exploiting unrestricted natural perturbations present severe security risks to deep learning systems, yet their transferability across models remains limited due to distribution mismatches between generated adversarial…

计算机视觉与模式识别 · 计算机科学 2025-05-23 Yuhao Xue , Zhifei Zhang , Xinyang Jiang , Yifei Shen , Junyao Gao , Wentao Gu , Jiale Zhao , Miaojing Shi , Cairong Zhao

Diffusion probabilistic models have been successfully used to generate data from noise. However, most diffusion models are computationally expensive and difficult to interpret with a lack of theoretical justification. Random feature models…

机器学习 · 统计学 2025-08-11 Esha Saha , Giang Tran

Network diffusion models are used to study disease transmission, information spread, technology adoption, and other socio-economic processes. We show that estimates of these diffusions are highly non-robust to mismeasurement. First, even…

计量经济学 · 经济学 2026-03-31 Arun G. Chandrasekhar , Paul Goldsmith-Pinkham , Tyler H. McCormick , Samuel Thau , Jerry Wei

Controlling memorization in diffusion models is critical for applications that require generated data to closely match the training distribution. Existing approaches mainly focus on data centric or model centric modifications, treating the…

机器学习 · 计算机科学 2026-01-30 Thuy Phuong Vu , Mai Viet Hoang Do , Minhhuy Le , Dinh-Cuong Hoang , Phan Xuan Tan

Large language models are increasingly deployed across diverse applications. This often includes tasks LLMs have not encountered during training. This implies that enumerating and obtaining the high-quality training data for all tasks is…

计算与语言 · 计算机科学 2025-11-11 Shambhavi Krishna , Atharva Naik , Chaitali Agarwal , Sudharshan Govindan , Taesung Lee , Haw-Shiuan Chang

Domain Adaptation (DA) has recently received significant attention due to its potential to adapt a learning model across source and target domains with mismatched distributions. Since DA methods rely exclusively on the given source and…

机器学习 · 统计学 2022-11-01 Akram S. Awad , George K. Atia

There is growing evidence that converting targets to soft targets in supervised learning can provide considerable gains in performance. Much of this work has considered classification, converting hard zero-one values to soft labels---such…

机器学习 · 统计学 2018-06-13 Ehsan Imani , Martha White

Training diffusion models requires large datasets. However, acquiring large volumes of high-quality data can be challenging, for example, collecting large numbers of high-resolution images and long videos. On the other hand, there are many…

计算机视觉与模式识别 · 计算机科学 2025-05-20 Xudong Ma

Cross-modal image translation remains brittle and inefficient. Standard diffusion approaches often rely on a single, global linear transfer between domains. We find that this shortcut forces the sampler to traverse off-manifold, high-cost…

计算机视觉与模式识别 · 计算机科学 2026-02-03 Zihao Wang , Yuzhou Chen , Shaogang Ren

Online learning with streaming data in a distributed and collaborative manner can be useful in a wide range of applications. This topic has been receiving considerable attention in recent years with emphasis on both single-task and…

多智能体系统 · 计算机科学 2017-04-26 Jie Chen , Cédric Richard , Ali H. Sayed

While the mathematical foundations of score-based generative models are increasingly well understood for unconstrained Euclidean spaces, many practical applications involve data restricted to bounded domains. This paper provides a…

统计理论 · 数学 2026-03-26 Asbjørn Holk , Claudia Strauch , Lukas Trottner

While diffusion models have achieved great success in generating continuous signals such as images and audio, it remains elusive for diffusion models in learning discrete sequence data like natural languages. Although recent advances…

计算与语言 · 计算机科学 2024-05-02 Jiasheng Ye , Zaixiang Zheng , Yu Bao , Lihua Qian , Mingxuan Wang
‹ 上一页 1 8 9 10 下一页 ›