图像生成 - arXivDaily 专题

1910.12027 2020-02-20 cs.LG cs.CV stat.ML 57%

Consistency Regularization for Generative Adversarial Networks

Han Zhang, Zizhao Zhang, Augustus Odena, Honglak Lee

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

Comments ICLR2020

2001.07321 2020-01-22 cs.CV cs.LG 57%

Neural Style Difference Transfer and Its Application to Font Generation

Gantugs Atarsaikhan, Brian Kenji Iwana, Seiichi Uchida

专题命中个性化与一致性：image editing(abstract)；分类 cs.CV

Comments Submitted to DAS2020

1910.08705 2019-10-22 eess.IV cs.CV 57%

Attention Guided Metal Artifact Correction in MRI using Deep Neural Networks

Jee Won Kim, Kinam Kwon, Byungjai Kim, HyunWook Park

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

Comments 6 pages, 5 figures

Journal ref ICCV 2019 Workshop on Interpreting and Explaining Visual Artificial Intelligence Models

1904.09261 2019-09-04 cs.CV 57%

Fashion++: Minimal Edits for Outfit Improvement

Wei-Lin Hsiao, Isay Katsman, Chao-Yuan Wu, Devi Parikh, Kristen Grauman

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

Comments accepted to ICCV 2019

1802.09655 2019-03-19 cs.CV 57%

Translating and Segmenting Multimodal Medical Volumes with Cycle- and Shape-Consistency Generative Adversarial Network

Zizhao Zhang, Lin Yang, Yefeng Zheng

专题命中个性化与一致性：image synthesis(abstract)；分类 cs.CV

Comments CVPR2018 (a small revision in v2)

1812.01288 2018-12-05 cs.CV 57%

FaceFeat-GAN: a Two-Stage Approach for Identity-Preserving Face Synthesis

Yujun Shen, Bolei Zhou, Ping Luo, Xiaoou Tang

专题命中个性化与一致性：image synthesis(abstract)；分类 cs.CV

Comments 12 pages and 6 figures

1811.05621 2018-11-15 cs.CV 57%

Style and Content Disentanglement in Generative Adversarial Networks

Hadi Kazemi, Seyed Mehdi Iranmanesh, Nasser M. Nasrabadi

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

Comments WACV 2019

1807.08772 2018-07-25 cs.CV 57%

Identity Preserving Face Completion for Large Ocular Region Occlusion

Yajie Zhao, Weikai Chen, Jun Xing, Xiaoming Li, Zach Bessinger, Fuchang Liu, Wangmeng Zuo, Ruigang Yang

专题命中个性化与一致性：inpainting(abstract)；分类 cs.CV

Comments 12 pages,9 figures

Journal ref The British Machine Vision Conference (BMVC) 2018

1804.04779 2018-07-25 cs.CV cs.CR 57%

A Hybrid Model for Identity Obfuscation by Face Replacement

Qianru Sun, Ayush Tewari, Weipeng Xu, Mario Fritz, Christian Theobalt, Bernt Schiele

专题命中个性化与一致性：image synthesis(abstract)；分类 cs.CV

Comments ECCV'18, camera-ready version

1712.03742 2018-05-23 cs.CV cs.RO 57%

Domain Adaptation Using Adversarial Learning for Autonomous Navigation

Jaeyoon Yoo, Yongjun Hong, YungKyun Noh, Sungroh Yoon

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

1805.07997 2018-05-22 cs.CV stat.ML 57%

Anime Style Space Exploration Using Metric Learning and Generative Adversarial Networks

Sitao Xiang, Hao Li

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

1805.05269 2018-05-15 cs.CV 57%

Normal Similarity Network for Generative Modelling

Jay Nandy, Wynne Hsu, Mong Li Lee

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

1804.06078 2018-04-18 cs.CV cs.AI cs.IT math.IT 57%

Cross-Domain Adversarial Auto-Encoder

Haodi Hou, Jing Huo, Yang Gao

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

Comments Under review as a conference paper of KDD 2018

1711.06148 2018-03-29 cs.CV cs.LG 57%

Learning Compositional Visual Concepts with Mutual Consistency

Yunye Gong, Srikrishna Karanam, Ziyan Wu, Kuan-Chuan Peng, Jan Ernst, Peter C. Doerschuk

专题命中个性化与一致性：image synthesis(abstract)；分类 cs.CV

Comments 10 pages, 8 figures, 4 tables, CVPR 2018

1802.08057 2018-02-23 cs.CV 57%

MagnifyMe: Aiding Cross Resolution Face Recognition via Identity Aware Synthesis

Maneet Singh, Shruti Nagpal, Richa Singh, Mayank Vatsa, Angshul Majumdar

专题命中个性化与一致性：image synthesis(abstract)；分类 cs.CV

1705.08824 2017-11-30 cs.CV 57%

From source to target and back: symmetric bi-directional adaptive GAN

Paolo Russo, Fabio Maria Carlucci, Tatiana Tommasi, Barbara Caputo

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

1701.02096 2017-11-07 cs.CV 57%

Improved Texture Networks: Maximizing Quality and Diversity in Feed-forward Stylization and Texture Synthesis

Dmitry Ulyanov, Andrea Vedaldi, Victor Lempitsky

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

1704.01745 2017-04-07 cs.CV 57%

How to Make an Image More Memorable? A Deep Style Transfer Approach

Aliaksandr Siarohin, Gloria Zen, Cveta Majtanovic, Xavier Alameda-Pineda, Elisa Ricci, Nicu Sebe

专题命中个性化与一致性：image synthesis(abstract)；分类 cs.CV

Comments Accepted at ACM ICMR 2017

1603.05631 2016-07-27 cs.CV 57%

Generative Image Modeling using Style and Structure Adversarial Networks

Xiaolong Wang, Abhinav Gupta

专题命中个性化与一致性：image generation(abstract)；分类 cs.CV

1212.5376 2012-12-24 math.AP 56%

A basic identity for Kolmogorov operators in the space of continuous functions related to RDEs with multiplicative noise

Sandra Cerrai, Giuseppe Da Prato

专题命中个性化与一致性：diffusion(abstract,comments)

Comments Key words: Stochastic reaction-diffusion equations, Kolmogorov operators, Poincaré inequality, spectral gap, Sobolev spaces in infinite dimensional spaces

2607.05011 2026-08-24 q-fin.PR q-fin.MF q-fin.TR 版本更新 50%

Reaction-boundary variance and adjoint-consistent local-volatility projection

反应边界方差与伴随一致的局部波动率投影

Chris Angstmann, Tim Gebbie

专题命中个性化与一致性：diffusion(abstract)

AI总结推导潜在订单簿反应边界的运行时间方差核，分离日历时间波动率模型中常合并的三个对象，给出确定性活动时钟的局部波动率投影，指出伴随一致性是对运行到日历时间投影的现实约束。

Comments 13 pages, 12 figures, 1 table; expanded reaction-boundary volatility-surface analysis, clarified adjoint-consistency and clock-projection conditions, and minor reference corrections; Code reproducibility see: this https URL (https://github.com/timgebbie/reaction-boundary-vol-surface-reproducibility)

2606.22361 2026-08-20 cs.CL cs.AI 版本更新 50%

First-Token Broadcasters: Mechanistic Origins of Language Identity and Distributed Robustness in Transformers

首个令牌广播者：Transformer中语言身份与分布式鲁棒性的机制起源

Arjun Pillai, Christian Hoang, Anjelo Jann Laroza

机构 * Irvington High School（欧文顿高中）； GenAI4E ； Mapua University（马普阿大学）

专题命中个性化与一致性：diffusion(abstract)

AI总结通过因果干预方法LIHA，发现Transformer中少量注意力头（如GPT-2的L6H1）持续关注首个令牌并广播语言信号，且消融后补偿呈现层级前馈模式；指令微调将语言身份电路重组至早期层。

Comments Under review at Interp4Discovery @ NeurIPS 2026

2608.13925 2026-08-17 cs.LG cs.AI cs.CL 新提交 50%

CForce: Boosting Parallel Decoding for dLLMs via Consistency Forcing

CForce：通过一致性强制提升扩散大语言模型（dLLMs）的并行解码性能

Yuji Ren, Chenkai Xu, Zhuocheng Gong, Jianguo Li, Zhijie Deng

机构 * Shanghai Jiao Tong University（上海交通大学）； Ant Group（蚂蚁集团）

专题命中个性化与一致性：diffusion(abstract)

AI总结本文提出CForce方法，通过一致性强制提升dLLMs的并行解码性能，在LLaDA模型上实验证实其在高并行解码预算下可优化速度-质量权衡。

2509.23548 2026-08-12 cs.LG cs.AI 50%

Disentanglement of Variations with Multimodal Generative Modeling

Yijie Zhang, Yiyang Shen, Weiran Wang

机构 * Department of Computer Science University of Iowa（计算机科学系印第安纳大学）

专题命中个性化与一致性：diffusion(abstract)

Comments 22 pages, 14 figures, 7 tables

Journal ref International Conference on Learning Representations (ICLR), 2026

2608.02828 2026-08-05 math.ST math.PR q-fin.ST stat.TH 新提交 50%

Proper-score observation-driven filters: local geometry, estimation, and continuous-time limits

恰当评分观测驱动滤波器：局部几何、估计与连续时间极限

Giulia Livieri, Gianluca Palmari

专题命中个性化与一致性：diffusion(abstract)

AI总结该研究提出基于恰当评分规则的观测驱动滤波器，推导其局部动力学、渐近性质与连续时间极限，通过实验验证其在金融收益数据中的应用，为误设下的观测驱动滤波提供准则框架。

Comments 88 pages, 12 figures, 25 tables

2607.27512 2026-07-31 cs.CL cs.MA cs.SI 新提交 50%

Belief Coevolution in a Social Network of Generalist and Specialist Large Language Models

通用型与专家型大语言模型组成的社交网络中的信念协同演化

Germans Savcisens, Samantha Dies, Courtney Maynard, Tina Eliassi-Rad

机构 * Khoury College of Computer Sciences, Northeastern University（东北大学库尔计算机科学学院）； Network Science Institute, Northeastern University（东北大学网络科学研究所）； Santa Fe Institute（圣塔菲研究所）

专题命中个性化与一致性：diffusion(abstract)

AI总结该研究提出CoevolveSim框架，模拟通用型与专家型LLM组成的社交网络中信念扩散，发现专家型LLM可显著提升共识转变幅度，模拟多智能体LLM系统信念扩散需多样化底层LLM。

Comments 33 pages (14 pages of main text), 7 figures, 14 tables

2509.02240 2026-07-30 cond-mat.mtrl-sci physics.app-ph 50%

Improving atomic force microscopy structure discovery via style-translation

Jie Huang, Niko Oinonen, Fabio Priante, Filippo Federici Canova, Lauri Kurki, Chen Xu, Adam S. Foster

专题命中个性化与一致性：image generation(abstract)

Comments 14 pages, 7 figures

Journal ref npj Computational Materials (2026)

2607.22637 2026-07-28 cs.AI cs.IT math.IT 新提交 50%

Fast Cross-Scenario Adaptation of CSI Models via Channel Conditional Parameter Generation

通过信道条件参数生成实现CSI模型的快速跨场景自适应

Xudong Zou, Siyu Wu, Zunlei Feng, Jie Song, Yuanyu Wan, Mingli Song, Jiacong Hu

机构 * School of Software Technology, Zhejiang University（浙江大学软件学院）； State Key Laboratory of Blockchain and Data Security, Zhejiang University Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security（浙江大学杭州高新技术产业开发区（滨江）区块链与数据安全国家重点实验室）

专题命中个性化与一致性：diffusion(abstract)

AI总结研究针对CSI模型在不同场景适应性差的问题，提出CCPG方法，通过组件冻结实验识别瓶颈，生成轻量级LoRA权重，经多种技术处理实现快速跨场景自适应，实验表明该方法能高效部署CSI模型用于6G通信。

2607.21620 2026-07-27 cs.AI 新提交 50%

From Profiles to Steering Vectors: Global Sparse Priors and Local Semantic Calibration for Personalized Text Generation

从概要到导向向量：用于个性化文本生成的全局稀疏先验和局部语义校准

Liuji Chen, Zeyu Zhang, Xinyuan Zhang, Shuai Nie, Qiang Liu, Shu Wu, Liang Wang

机构 * NLPR, Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所模式识别国家重点实验室）； Xiaomi（小米）

专题命中个性化与一致性：personalized generation(abstract)

AI总结研究个性化文本生成问题，提出无训练框架GLASS，通过全局-局部激活导向与稀疏先验实现。利用稀疏自动编码器提取全局先验、构建局部向量，推理时注入模型层。实验证明其性能优于基线，能更好分离风格与语义信息。

Comments Under review

2607.20127 2026-07-23 cs.CL 新提交 50%

Back to Back with a Copy: A Computational Analysis of AI-Generated Visual Contemporary Art Pastiches

与仿作并肩：人工智能生成的视觉当代艺术模仿作品的计算分析

Anca Dinu, Andreiana Mihail, Andra-Maria Florescu, Claudiu Creanga, Liviu Dinu

机构 * University of Bucharest（布加勒斯特大学）

专题命中个性化与一致性：image generation(abstract)

AI总结本文研究更新的生成模型模仿当代艺术作品的能力及不同语言模型中风格评估的多维度一致性。利用五个计算机视觉模型，通过余弦距离捕捉多种特征，对比新旧模型。结果显示新模型语义对齐改进、多样性增加，但浅层特征表现稍弱，还通过艺术家反馈进行情境化。

1. 个性化与一致性 1272 篇

Consistency Regularization for Generative Adversarial Networks

Neural Style Difference Transfer and Its Application to Font Generation

Attention Guided Metal Artifact Correction in MRI using Deep Neural Networks

Fashion++: Minimal Edits for Outfit Improvement

Translating and Segmenting Multimodal Medical Volumes with Cycle- and Shape-Consistency Generative Adversarial Network

FaceFeat-GAN: a Two-Stage Approach for Identity-Preserving Face Synthesis

Style and Content Disentanglement in Generative Adversarial Networks

Identity Preserving Face Completion for Large Ocular Region Occlusion

A Hybrid Model for Identity Obfuscation by Face Replacement

Domain Adaptation Using Adversarial Learning for Autonomous Navigation

Anime Style Space Exploration Using Metric Learning and Generative Adversarial Networks

Normal Similarity Network for Generative Modelling

Cross-Domain Adversarial Auto-Encoder

Learning Compositional Visual Concepts with Mutual Consistency

MagnifyMe: Aiding Cross Resolution Face Recognition via Identity Aware Synthesis

From source to target and back: symmetric bi-directional adaptive GAN

Improved Texture Networks: Maximizing Quality and Diversity in Feed-forward Stylization and Texture Synthesis

How to Make an Image More Memorable? A Deep Style Transfer Approach

Generative Image Modeling using Style and Structure Adversarial Networks

A basic identity for Kolmogorov operators in the space of continuous functions related to RDEs with multiplicative noise

Reaction-boundary variance and adjoint-consistent local-volatility projection

First-Token Broadcasters: Mechanistic Origins of Language Identity and Distributed Robustness in Transformers

CForce: Boosting Parallel Decoding for dLLMs via Consistency Forcing

Disentanglement of Variations with Multimodal Generative Modeling

Proper-score observation-driven filters: local geometry, estimation, and continuous-time limits

Belief Coevolution in a Social Network of Generalist and Specialist Large Language Models

Improving atomic force microscopy structure discovery via style-translation

Fast Cross-Scenario Adaptation of CSI Models via Channel Conditional Parameter Generation

From Profiles to Steering Vectors: Global Sparse Priors and Local Semantic Calibration for Personalized Text Generation

Back to Back with a Copy: A Computational Analysis of AI-Generated Visual Contemporary Art Pastiches