图像生成

2507.22746 2025-08-04 cs.SD cs.CL eess.AS 50%

Next Tokens Denoising for Speech Synthesis

Yanqing Liu, Ruiqing Xue, Chong Zhang, Yufei Liu, Gang Wang, Bohan Li, Yao Qian, Lei He, Shujie Liu, Sheng Zhao

机构 * Microsoft（微软）

专题命中扩散模型：diffusion(abstract)

2507.21053 2025-08-04 cs.LG cs.RO 50%

Flow Matching Policy Gradients

David McAllister, Songwei Ge, Brent Yi, Chung Min Kim, Ethan Weber, Hongsuk Choi, Haiwen Feng, Angjoo Kanazawa

机构 * UC Berkeley（伯克利大学）； Max Planck Institute for Intelligent Systems（智能系统马克斯·普朗克研究所）

专题命中扩散模型：diffusion(abstract)

Comments See our blog post at https://flowreinforce.github.io

2507.20403 2025-08-04 econ.TH cs.LG 50%

A General Framework for Estimating Preferences Using Response Time Data

Federico Echenique, Alireza Fallah, Michael I. Jordan

机构 * Department of Economics, University of California, Berkeley（加州大学伯克利分校经济学系）； Department of Computer Science, Rice University（里德大学计算机科学系）； Departments of Electrical Engineering and Computer Sciences and Statistics, University of California, Berkeley（加州大学伯克利分校电子工程与计算机科学及统计学系；巴黎Inria）； Inria Paris

专题命中扩散模型：diffusion(abstract)

2507.18004 2025-08-04 cs.AI 50%

E.A.R.T.H.: Structuring Creative Evolution through Model Error in Generative AI

Yusen Peng, Shuhua Mao

机构 * University of Warwick（沃里克大学）； Wuhan University of Technology（武汉理工大学）

专题命中扩散模型：diffusion(abstract)

Comments 44 pages,11 figures

2503.03778 2025-08-04 eess.IV q-bio.TO 50%

Generating Novel Brain Morphology by Deforming Learned Templates

Alan Q. Wang, Fangrui Huang, Bailey Trang, Wei Peng, Mohammad Abbasi, Kilian Pohl, Mert Sabuncu, Ehsan Adeli

专题命中扩散模型：diffusion(abstract)

Comments Provisional Acceptance at MICCAI 2025

2502.01533 2025-08-04 cs.LG cs.AI q-bio.BM 50%

Transformers trained on proteins can learn to attend to Euclidean distance

Isaac Ellmen, Constantin Schneider, Matthew I. J. Raybould, Charlotte M. Deane

专题命中扩散模型：diffusion(abstract)

Journal ref Transactions on Machine Learning Research (2025) https://openreview.net/forum?id=mU59bDyqqv

2502.01300 2025-08-04 physics.plasm-ph astro-ph.SR physics.space-ph 50%

Magnetic Field Amplification and Reconstruction in Rotating Astrophysical Plasmas: Verifying the Roles of $α$ and $β$ in Dynamo Action

Kiwan Park

专题命中扩散模型：diffusion(abstract)

Comments 26 pages, 8 figures, submitted

2411.01355 2025-08-04 cond-mat.stat-mech cond-mat.soft math-ph math.MP math.PR 50%

Spatio-temporal fluctuations in the passive and active Riesz gas on the circle

Leo Touzo, Pierre Le Doussal, Gregory Schehr

专题命中扩散模型：diffusion(abstract)

Comments 76 pages, 13 figures

Journal ref J. Stat. Phys. 192, 79 (2025)

2211.08242 2025-08-04 math.PR 50%

Exponential ergodicity of stochastic heat equations with Hölder coefficients

Yi Han

专题命中扩散模型：diffusion(abstract)

Comments 42 pages. To appear in SIAM Journal on Mathematical Analysis

Journal ref SIAM Journal on Mathematical Analysis Volume 57 Issue 4 August 2025 Pages: 4220 - 4263

2508.00272 2025-08-04 cs.CV 70%

Towards Robust Semantic Correspondence: A Benchmark and Insights

Wenyue Chong

专题命中可控生成：diffusion(abstract)；image editing(abstract)；分类 cs.CV

2411.15867 2025-08-04 cs.CV 70%

PanoLlama: Generating Endless and Coherent Panoramas with Next-Token-Prediction LLMs

Teng Zhou, Xiaoyu Zhang, Yongchuan Tang

机构 * College of Computer Science and Technology（计算机科学与技术学院）

专题命中可控生成：image generation(abstract)；diffusion(abstract)；分类 cs.CV

2508.00299 2025-08-04 cs.CV cs.AI cs.RO 57%

Controllable Pedestrian Video Editing for Multi-View Driving Scenarios via Motion Sequence

Danzhen Fu, Jiagao Hu, Daiguo Zhou, Fei Wang, Zepeng Wang, Wenhua Liao

机构 * MiLM Plus, Xiaomi Inc.（小米公司）

专题命中可控生成：inpainting(abstract)；分类 cs.CV

Comments ICCV 2025 Workshop (HiGen)

2312.12699 2025-08-04 math.NA cs.NA 50%

Propagation of chaos in infinite horizon and numerical stability for stochastic McKean-Vlasov equations

Zhuoqi Liu, Shuaibin Gao, Chenggui Yuan, Qian Guo

专题命中可控生成：diffusion(abstract)

2309.12014 2025-08-04 q-fin.RM math.OC q-fin.MF 50%

Singular Control in a Cash Management Model with Ambiguity

Arnon Archankul, Giorgio Ferrari, Tobias Hellmann, Jacco J. J. Thijssen

专题命中可控生成：diffusion(abstract)

2508.00427 2025-08-04 cs.CV cs.AI 83%

Contact-Aware Amodal Completion for Human-Object Interaction via Multi-Regional Inpainting

Seunggeun Chi, Enna Sachdeva, Pin-Hao Huang, Kwonjoon Lee

机构 * Purdue University West Lafayette（普渡大学西拉法叶分校）； Honda Research Institute USA（本田美国研究院）

专题命中图像修复：inpainting(title,abstract)；diffusion(abstract)；分类 cs.CV

Comments ICCV 2025 (Highlight)

2508.00418 2025-08-04 cs.CV eess.IV 79%

IN2OUT: Fine-Tuning Video Inpainting Model for Video Outpainting Using Hierarchical Discriminator

Sangwoo Youn, Minji Lee, Nokap Tony Park, Yeonggyoo Jeon, Taeyoung Na

机构 * KAIST（韩国科学技术院）； Columbia University（哥伦比亚大学）； SK Telecom（SK电信）

专题命中图像修复：inpainting(title,abstract)；分类 cs.CV

Comments ICIP 2025. Code: https://github.com/sang-w00/IN2OUT

2508.00397 2025-08-04 cs.CV 57%

Video Forgery Detection with Optical Flow Residuals and Spatial-Temporal Consistency

Xi Xue, Kunio Suzuki, Nabarun Goswami, Takuya Shintate

机构 * NABLAS Inc.（NABLAS公司）

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

2508.00428 2025-08-04 cs.GR cs.HC 57%

Sel3DCraft: Interactive Visual Prompts for User-Friendly Text-to-3D Generation

Nan Xiang, Tianyi Liang, Haiwen Huang, Shiqi Jiang, Hao Huang, Yifei Huang, Liangyu Chen, Changbo Wang, Chenhui Li

专题命中图像生成评测：text-to-image(abstract)；分类 cs.GR

Comments IEEE VIS VAST 2025 ACM 2012 CCS - Human-centered computing, Visualization, Visualization design and evaluation methods

2508.00260 2025-08-04 cs.CV cs.MM 81%

Instruction-Grounded Visual Projectors for Continual Learning of Generative Vision-Language Models

Hyundong Jin, Hyung Jin Chang, Eunwoo Kim

机构 * School of Computer Science and Engineering, Chung-Ang University（Chung-Ang 大学计算机科学与工程学院）； School of Computer Science, University of Birmingham（布拉德福德大学计算机科学学院）

专题命中其他图像生成：generative vision(title,abstract)；分类 cs.CV、cs.MM

Comments Accepted to ICCV 2025

1. 扩散模型 34 篇

Next Tokens Denoising for Speech Synthesis

Flow Matching Policy Gradients

A General Framework for Estimating Preferences Using Response Time Data

E.A.R.T.H.: Structuring Creative Evolution through Model Error in Generative AI

Generating Novel Brain Morphology by Deforming Learned Templates

Transformers trained on proteins can learn to attend to Euclidean distance

Magnetic Field Amplification and Reconstruction in Rotating Astrophysical Plasmas: Verifying the Roles of $α$ and $β$ in Dynamo Action

Spatio-temporal fluctuations in the passive and active Riesz gas on the circle

Exponential ergodicity of stochastic heat equations with Hölder coefficients

2. 可控生成 5 篇

Towards Robust Semantic Correspondence: A Benchmark and Insights

PanoLlama: Generating Endless and Coherent Panoramas with Next-Token-Prediction LLMs

Controllable Pedestrian Video Editing for Multi-View Driving Scenarios via Motion Sequence

Propagation of chaos in infinite horizon and numerical stability for stochastic McKean-Vlasov equations

Singular Control in a Cash Management Model with Ambiguity

3. 图像修复 2 篇

Contact-Aware Amodal Completion for Human-Object Interaction via Multi-Regional Inpainting

IN2OUT: Fine-Tuning Video Inpainting Model for Video Outpainting Using Hierarchical Discriminator

4. 个性化与一致性 1 篇

Video Forgery Detection with Optical Flow Residuals and Spatial-Temporal Consistency

5. 图像生成评测 1 篇

Sel3DCraft: Interactive Visual Prompts for User-Friendly Text-to-3D Generation

6. 其他图像生成 1 篇

Instruction-Grounded Visual Projectors for Continual Learning of Generative Vision-Language Models