图像生成 - arXivDaily 专题

2207.10271 2022-07-29 cs.CV 79%

DeltaGAN: Towards Diverse Few-shot Image Generation with Sample-Specific Delta

Yan Hong, Li Niu, Jianfu Zhang, Liqing Zhang

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

Comments I want to withdraw this version and use it to update the previous version at arXiv:2009.08753

2204.11425 2022-05-11 eess.IV cs.CV 79%

BCI: Breast Cancer Immunohistochemical Image Generation through Pyramid Pix2pix

Shengjie Liu, Chuang Zhu, Feng Xu, Xinyu Jia, Zhongyue Shi, Mulan Jin

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

Comments Accepted by CVPR2022 Workshop

2112.09809 2022-05-05 cs.CV 79%

A Streaming Volumetric Image Generation Framework for Development and Evaluation of Out-of-Core Methods

Dominik Drees, Xiaoyi Jiang

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

2105.05332 2022-04-26 cs.CV 79%

The DEVIL is in the Details: A Diagnostic Evaluation Benchmark for Video Inpainting

Ryan Szeto, Jason J. Corso

专题命中图像生成评测：inpainting(title,abstract)；分类 cs.CV

Comments Accepted to CVPR 2022

2004.12361 2021-02-09 cs.CV cs.LG eess.IV 79%

Evaluation Metrics for Conditional Image Generation

Yaniv Benny, Tomer Galanti, Sagie Benaim, Lior Wolf

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

Comments To be published in "INTERNATIONAL JOURNAL OF COMPUTER VISION"

2003.07449 2020-12-04 cs.CV cs.LG eess.IV 79%

Object-Centric Image Generation from Layouts

Tristan Sylvain, Pengchuan Zhang, Yoshua Bengio, R Devon Hjelm, Shikhar Sharma

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

Comments AAAI 2021

2005.06968 2020-09-16 cs.LG cs.CL cs.CV 79%

S2IGAN: Speech-to-Image Generation via Adversarial Learning

Xinsheng Wang, Tingting Qiao, Jihua Zhu, Alan Hanjalic, Odette Scharenborg

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

Comments Accepted to Interspeech2020

2003.11774 2020-04-01 cs.CV cs.LG eess.IV 79%

Image Generation Via Minimizing Fréchet Distance in Discriminator Feature Space

Khoa D. Doan, Saurav Manchanda, Fengjiao Wang, Sathiya Keerthi, Avradeep Bhowmik, Chandan K. Reddy

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

1912.05028 2019-12-12 cs.CV cs.LG eess.IV 79%

cFineGAN: Unsupervised multi-conditional fine-grained image generation

Gunjan Aggarwal, Abhishek Sinha

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

Comments Accepted at NeurIPS Workshop on Machine Learning for Creativity and Design 3.0

1901.03762 2019-01-17 cs.CV 79%

Using Scene Graph Context to Improve Image Generation

Subarna Tripathi, Anahita Bhiwandiwalla, Alexei Bastidas, Hanlin Tang

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

Comments arXiv admin note: text overlap with arXiv:1804.01622 by other authors

1901.02199 2019-01-09 cs.LG cs.CV stat.ML 79%

FIGR: Few-shot Image Generation with Reptile

Louis Clouâtre, Marc Demers

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

Comments 9 pages

1804.10323 2018-04-30 cs.CV 79%

Adversarial Training of Variational Auto-encoders for High Fidelity Image Generation

Salman H. Khan, Munawar Hayat, Nick Barnes

专题命中图像生成评测：image generation(title,abstract)；分类 cs.CV

2607.24763 2026-07-29 cs.AI 新提交 78%

CaRE Compute-aware Remasking Evaluation Protocol for Masked Diffusion Language Models

用于掩码扩散语言模型的计算感知重掩码评估协议

Yash Shah, Abhijit Chakraborty, Vivek Gupta

机构 * Arizona State University（亚利桑那州立大学）； MongoDB（MongoDB公司）

专题命中图像生成评测：diffusion(title,abstract)

AI总结研究针对掩码扩散语言模型评估标准不完善的问题，提出计算感知评估框架CaRE，通过标准化函数评估实际数量等方法审核重掩码策略，应用于多种模型和数据集，揭示了温度、计算匹配等因素对评估的影响，发布相关内容确保评估可重复可比。

Comments updated version

2607.13120 2026-07-16 cs.LG cs.AI 新提交 78%

CoDiffGRN: Rethinking Gene Regulatory Network Inference via the BEELINE-KGC Benchmark and Co-evolutionary Discrete Diffusion

CoDiffGRN：通过BEELINE-KGC基准和协同进化离散扩散重新思考基因调控网络推理

Jiaze Song, Runhao Zhao, Minghao Xu, Bin Cui, Wentao Zhang

机构 * Peking University（北京大学）； National University of Defense Technology（国防科技大学）

专题命中图像生成评测：diffusion(title,abstract)

AI总结该研究针对单细胞转录组数据推断基因调控网络问题，提出CoDiffGRN方法，将GRN推理转化为归纳图完成问题，引入新基准和协同进化离散扩散框架，实验表明其在新调控发现上性能优越，优于现有方法。

Comments 19 pages, 6 figures

2607.06609 2026-07-09 cs.LG cs.AI 新提交 78%

D2PO: Optimizing Diffusion Samplers via Dynamic Preference

D2PO：通过动态偏好优化扩散采样器

Jinkyu Kim, Jinyoung Choi, Bohyung Han

机构 * Seoul National University（首尔国立大学）； Ulsan National Institute of Science and Technology（蔚山国立科学技术院）

专题命中图像生成评测：diffusion(title,abstract)

AI总结研究针对现有框架局限，提出D2PO框架优化扩散采样策略。核心方法是将其转化为基于偏好的对齐问题，利用DPO框架并建模为能量模型，引入新能量公式和动态偏好。主要贡献是使采样器与感知质量更忠实对齐，性能优于传统调度器。

Comments Accepted to ECCV 2026

2607.01709 2026-07-03 cs.AI cs.LG 新提交 78%

COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows

COMFYCLAW：面向图像生成工作流的自进化技能工具包

Zongxia Li, Dawei Liu, Fuxiao Liu, Yuhang Zhou, Xiyang Wu, Jingxi Chen, Jing Xie, Xiaomin Wu, Lichao Sun

机构 * University of Maryland（马里兰大学）； University of Pennsylvania（宾夕法尼亚大学）； Nvidia（英伟达）； Lehigh University（里海大学）

专题命中图像生成评测：image generation(title,abstract)

AI总结提出COMFYCLAW框架，通过类型化图编辑、区域级VLM验证器和渐进式技能库进化，提升ComfyUI工作流构建的代理可靠性和性能，在多个基准上取得最佳平均评分。

2605.07496 2026-06-08 cs.RO 版本更新 78%

PathPainter: Transferring the Generalization Ability of Image Generation Models to Embodied Navigation

PathPainter：将图像生成模型的泛化能力迁移至具身导航

Yijin Wang, Yuru Tian, Xijie Huang, Weiqi Gai, Mo Zhu, Xin Zhou, Yuze Wu, Fei Gao

机构 * Tsinghua University（清华大学）

专题命中图像生成评测：image generation(title,abstract)

AI总结提出利用鸟瞰图作为全局先验的导航系统，通过图像生成模型理解自然语言意图并生成可通行掩码，结合跨视图定位消除里程计漂移，在无人机平台上完成160米室外长距离导航。

Comments Work in the progress. 16 pages, 13 figures

2604.13413 2026-04-16 cs.LG 78%

Dataset-Level Metrics Attenuate Non-Determinism: A Fine-Grained Non-Determinism Evaluation in Diffusion Language Models

数据层面指标削弱非确定性：扩散语言模型中的细粒度非确定性评估

Zhengyu Fang, Zhimeng Jiang, Huiyuan Chen, Xiaoge Zhang, Tianyi Li, Kaiyu Tang, Xiao Li, Jing Li

机构 * Department of Computer ； Data Sciences, Case Western Reserve University, Cleveland, USA ； Department of Computer Science \& Engineering, Texas A\&M University, College Station, USA ； Department of Biochemistry, Case Western Reserve University, Cleveland, USA ； Center for RNA Science ； Therapeutics, Case Western Reserve University, Cleveland, USA ； Department of Biomedical Engineering, Case Western Reserve University, Cleveland, USA

专题命中图像生成评测：diffusion(title,abstract)

AI总结本文研究了扩散语言模型中非确定性问题，指出数据层面指标限制了对模型行为变化的洞察，并提出基于样本级预测差异的细粒度评估方法，揭示了非确定性在代码生成中的高敏感性。

2604.02718 2026-04-06 cs.LG cs.CL 78%

Generative Frontiers: Why Evaluation Matters for Diffusion Language Models

生成前沿：为何评估对扩散语言模型至关重要

Patrick Pynadath, Jiaxin Shi, Ruqi Zhang

机构 * Purdue University（普渡大学）

专题命中图像生成评测：diffusion(title,abstract)

AI总结本文探讨了扩散语言模型评估方法的局限性，提出生成前沿作为评估生成质量的原理性方法，分析了生成困惑度和熵在KL散度中的作用。

2509.23348 2026-03-04 cs.LG 78%

Entering the Era of Discrete Diffusion Models: A Benchmark for Schrödinger Bridges and Entropic Optimal Transport

进入离散扩散模型的时代：Schrödinger桥和熵最优传输的基准

Xavier Aramayo Carrasco, Grigoriy Ksenofontov, Aleksei Leonov, Iaroslav Sergeevich Koshelev, Alexander Korotin

机构 * Applied AI Institute（应用人工智能研究所）； AI Foundation and Algorithm Lab（人工智能基金会与算法实验室）

专题命中图像生成评测：diffusion(title,abstract)

AI总结本文提出了一种针对离散空间的Schrödinger桥基准，通过构造具有已知解的概率分布对，实现了对SB方法在高维离散设置中性能的严格评估，并引入了新的SB算法DLightSB和DLightSB-M以及α-CSBM算法。

2511.00530 2025-11-04 cs.IR 78%

Listwise Preference Diffusion Optimization for User Behavior Trajectories Prediction

Hongtao Huang, Chengkai Huang, Junda Wu, Tong Yu, Julian McAuley, Lina Yao

专题命中图像生成评测：diffusion(title,abstract)

2505.21347 2025-10-28 cs.LG 78%

OVERT: A Benchmark for Over-Refusal Evaluation on Text-to-Image Models

Ziheng Cheng, Yixiao Huang, Hui Xu, Somayeh Sojoudi, Xuandong Zhao, Dawn Song, Song Mei

机构 * UC Berkeley（加州大学伯克利分校）

专题命中图像生成评测：text-to-image(title,abstract)

Comments NeurIPS 2025 (D&B Track)

2509.26063 2025-10-01 cs.IR 78%

Fading to Grow: Growing Preference Ratios via Preference Fading Discrete Diffusion for Recommendation

Guoqing Hu, An Zhang. Shuchang Liu, Wenyu Mao, Jiancan Wu, Xun Yang, Xiang Li, Lantao Hu, Han Li, Kun Gai, Xiang Wang

专题命中图像生成评测：diffusion(title,abstract)

Journal ref NeurIPS 2025

2502.11798 2025-07-22 cs.CR 78%

BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model

Weilin Lin, Nanjun Zhou, Yanyun Wang, Jianze Li, Hui Xiong, Li Liu

专题命中图像生成评测：diffusion(title,abstract)

2408.15501 2025-05-28 cs.LG cs.AI 78%

MODULI: Unlocking Preference Generalization via Diffusion Models for Offline Multi-Objective Reinforcement Learning

Yifu Yuan, Zhenrui Zheng, Zibin Dong, Jianye Hao

专题命中图像生成评测：diffusion(title,abstract)

Comments Accepted by ICML2025, code link: https://github.com/pickxiguapi/MODULI

2505.04263 2025-05-08 cs.LG cs.NA math.NA 78%

Physics-Informed DeepONets for drift-diffusion on metric graphs: simulation and parameter identification

Jan Blechschmidt, Tom-Christian Riemer, Max Winkler, Martin Stoll, Jan-F. Pietschmann

专题命中图像生成评测：diffusion(title,abstract)

2505.04146 2025-05-08 cs.CL cs.AI cs.CY 78%

Unmasking the Canvas: A Dynamic Benchmark for Image Generation Jailbreaking and LLM Content Safety

Variath Madhupal Gautham Nair, Vishal Varma Dantuluri

专题命中图像生成评测：image generation(title,abstract)

2503.18347 2025-03-25 cs.LG cs.AI cs.RO 78%

Latent Embedding Adaptation for Human Preference Alignment in Diffusion Planners

Wen Zheng Terence Ng, Jianda Chen, Yuan Xu, Tianwei Zhang

专题命中图像生成评测：diffusion(title,abstract)

Comments 8 pages

2502.03429 2025-02-06 cs.CL cs.AI 78%

On Fairness of Unified Multimodal Large Language Model for Image Generation

Ming Liu, Hao Chen, Jindong Wang, Liwen Wang, Bhiksha Raj Ramakrishnan, Wensheng Zhang

专题命中图像生成评测：image generation(title)；diffusion(abstract)

2309.09537 2025-01-16 cs.SI cs.AI 78%

Towards a performance characteristic curve for model evaluation: an application in information diffusion prediction

Wenjin Xie, Xiaomeng Wang, Radosław Michalski, Tao Jia

专题命中图像生成评测：diffusion(title,abstract)

1. 图像生成评测 1204 篇

DeltaGAN: Towards Diverse Few-shot Image Generation with Sample-Specific Delta

BCI: Breast Cancer Immunohistochemical Image Generation through Pyramid Pix2pix

A Streaming Volumetric Image Generation Framework for Development and Evaluation of Out-of-Core Methods

The DEVIL is in the Details: A Diagnostic Evaluation Benchmark for Video Inpainting

Evaluation Metrics for Conditional Image Generation

Object-Centric Image Generation from Layouts

S2IGAN: Speech-to-Image Generation via Adversarial Learning

Image Generation Via Minimizing Fréchet Distance in Discriminator Feature Space

cFineGAN: Unsupervised multi-conditional fine-grained image generation

Using Scene Graph Context to Improve Image Generation

FIGR: Few-shot Image Generation with Reptile

Adversarial Training of Variational Auto-encoders for High Fidelity Image Generation

CaRE Compute-aware Remasking Evaluation Protocol for Masked Diffusion Language Models

CoDiffGRN: Rethinking Gene Regulatory Network Inference via the BEELINE-KGC Benchmark and Co-evolutionary Discrete Diffusion

D2PO: Optimizing Diffusion Samplers via Dynamic Preference

COMFYCLAW: Self-Evolving Skill Harnesses for Image Generation Workflows

PathPainter: Transferring the Generalization Ability of Image Generation Models to Embodied Navigation

Dataset-Level Metrics Attenuate Non-Determinism: A Fine-Grained Non-Determinism Evaluation in Diffusion Language Models

Generative Frontiers: Why Evaluation Matters for Diffusion Language Models

Entering the Era of Discrete Diffusion Models: A Benchmark for Schrödinger Bridges and Entropic Optimal Transport

Listwise Preference Diffusion Optimization for User Behavior Trajectories Prediction

OVERT: A Benchmark for Over-Refusal Evaluation on Text-to-Image Models

Fading to Grow: Growing Preference Ratios via Preference Fading Discrete Diffusion for Recommendation

BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model

MODULI: Unlocking Preference Generalization via Diffusion Models for Offline Multi-Objective Reinforcement Learning

Physics-Informed DeepONets for drift-diffusion on metric graphs: simulation and parameter identification

Unmasking the Canvas: A Dynamic Benchmark for Image Generation Jailbreaking and LLM Content Safety

Latent Embedding Adaptation for Human Preference Alignment in Diffusion Planners

On Fairness of Unified Multimodal Large Language Model for Image Generation

Towards a performance characteristic curve for model evaluation: an application in information diffusion prediction