# Generative & Multimodal Models Domain: Artificial Intelligence · 273 papers indexed · newest 2026-10-28 Diffusion and flow-matching models, vision-language models, video generation and learned world models. Key concepts: diffusion model, flow matching, vision-language model, video generation, world model, multimodal large language model, text-to-image ## Most influential - **[Denoising Diffusion Probabilistic Models](https://space-kitty.com/papers/arxiv:2006.11239)**. Jonathan Ho, Jain, Ajay, Pieter Abbeel · 2020 (arXiv:2006.11239 · 8,714 citations). Source: > We present high quality image synthesis results using diffusion probabilistic models, a class of latent variable models inspired by considerations from nonequilibrium thermodynamics. Our best results are obtained by training on a weighted variational bound designed according to a novel connection between diffusion probabilistic models and denoising score matching with Langevin dynamics, and our models naturally admit a progressive lossy decompression scheme that can be interpreted as a generalization of autoregressive decoding. On the unconditional CIFAR10 dataset, we obtain an Inception score… - **[Diffusion Models in Vision: A Survey](https://space-kitty.com/papers/arxiv:2209.04747)**. Florinel-Alin Croitoru, Vlad Hondru, Radu Tudor Ionescu et al. · 2023 · IEEE Transactions on Pattern Analysis and Machine Intelligence (arXiv:2209.04747 · doi:10.1109/tpami.2023.3261988 · 1,909 citations). Source: > Denoising diffusion models represent a recent emerging topic in computer vision, demonstrating remarkable results in the area of generative modeling. A diffusion model is a deep generative model that is based on two stages, a forward diffusion stage and a reverse diffusion stage. In the forward diffusion stage, the input data is gradually perturbed over several steps by adding Gaussian noise. In the reverse stage, a model is tasked at recovering the original input data by learning to gradually reverse the diffusion process, step by step. Diffusion models are widely appreciated for the quality … - **[Denoising Diffusion Implicit Models](https://space-kitty.com/papers/arxiv:2010.02502)**. Jiaming Song, Chenlin Meng, Stefano Ermon · 2020 (arXiv:2010.02502 · 2,507 citations). Source: > Denoising diffusion probabilistic models (DDPMs) have achieved high quality image generation without adversarial training, yet they require simulating a Markov chain for many steps to produce a sample. To accelerate sampling, we present denoising diffusion implicit models (DDIMs), a more efficient class of iterative implicit probabilistic models with the same training procedure as DDPMs. In DDPMs, the generative process is defined as the reverse of a Markovian diffusion process. We construct a class of non-Markovian diffusion processes that lead to the same training objective, but whose revers… - **[Improved Denoising Diffusion Probabilistic Models](https://space-kitty.com/papers/arxiv:2102.09672)**. Alex Nichol, Prafulla Dhariwal · 2021 (arXiv:2102.09672 · 1,550 citations). Source: > Denoising diffusion probabilistic models (DDPM) are a class of generative models which have recently been shown to produce excellent samples. We show that with a few simple modifications, DDPMs can also achieve competitive log-likelihoods while maintaining high sample quality. Additionally, we find that learning variances of the reverse diffusion process allows sampling with an order of magnitude fewer forward passes with a negligible difference in sample quality, which is important for the practical deployment of these models. We additionally use precision and recall to compare how well DDPMs… - **[Diffusion models in medical imaging: A comprehensive survey](https://space-kitty.com/papers/doi:10.1016/j.media.2023.102846)**. Amirhossein Kazerouni, Ehsan Khodapanah Aghdam, Moein Heidari et al. · 2023 · Medical Image Analysis (doi:10.1016/j.media.2023.102846 · 776 citations). Source: - **[MotionDiffuse: Text-Driven Human Motion Generation With Diffusion Model](https://space-kitty.com/papers/arxiv:2208.15001)**. Mingyuan Zhang, Zhongang Cai, Liang Pan et al. · 2024 · IEEE Transactions on Pattern Analysis and Machine Intelligence (arXiv:2208.15001 · doi:10.1109/tpami.2024.3355414 · 437 citations). Source: > Human motion modeling is important for many modern graphics applications, which typically require professional skills. In order to remove the skill barriers for laymen, recent motion generation methods can directly generate human motions conditioned on natural languages. However, it remains challenging to achieve diverse and fine-grained motion generation with various text inputs. To address this problem, we propose MotionDiffuse, the first diffusion model-based text-driven motion generation framework, which demonstrates several desired properties over existing methods. 1) Probabilistic Mappin… - **[DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models](https://space-kitty.com/papers/arxiv:2211.01095)**. Cheng Lu, Yuhao Zhou, Fan Bao et al. · 2025 · Machine Intelligence Research (arXiv:2211.01095 · doi:10.1007/s11633-025-1562-4 · 280 citations). Source: > Diffusion probabilistic models (DPMs) have achieved impressive success in high-resolution image synthesis, especially in recent large-scale text-to-image generation applications. An essential technique for improving the sample quality of DPMs is guided sampling, which usually needs a large guidance scale to obtain the best sample quality. The commonly-used fast sampler for guided sampling is DDIM, a first-order diffusion ODE solver that generally needs 100 to 250 steps for high-quality samples. Although recent works propose dedicated high-order solvers and achieve a further speedup for samplin… - **[Restoring Vision in Adverse Weather Conditions With Patch-Based Denoising Diffusion Models](https://space-kitty.com/papers/arxiv:2207.14626)**. Ozan Özdenizci, Robert A. Legenstein · 2023 · IEEE Transactions on Pattern Analysis and Machine Intelligence (arXiv:2207.14626 · doi:10.1109/tpami.2023.3238179 · 475 citations). Source: > Image restoration under adverse weather conditions has been of significant interest for various computer vision applications. Recent successful methods rely on the current progress in deep neural network architectural designs (e.g., with vision transformers). Motivated by the recent progress achieved with state-of-the-art conditional generative models, we present a novel patch-based image restoration algorithm based on denoising diffusion probabilistic models. Our patch-based diffusion modeling approach enables size-agnostic image restoration by using a guided denoising process with smoothed n… - **[Denoising diffusion probabilistic models for 3D medical image generation](https://space-kitty.com/papers/doi:10.1038/s41598-023-34341-2)**. Firas Khader, Gustav Anton Müller‐Franzes, Soroosh Tayebi Arasteh et al. · 2023 · Scientific Reports (doi:10.1038/s41598-023-34341-2 · 343 citations). Source: > Recent advances in computer vision have shown promising results in image generation. Diffusion probabilistic models have generated realistic images from textual input, as demonstrated by DALL-E 2, Imagen, and Stable Diffusion. However, their use in medicine, where imaging data typically comprises three-dimensional volumes, has not been systematically evaluated. Synthetic images may play a crucial role in privacy-preserving artificial intelligence and can also be used to augment small datasets. We show that diffusion probabilistic models can synthesize high-quality medical data for magnetic res… > _Abstract excerpt by Firas Khader et al., CC-BY, https://doi.org/10.1038/s41598-023-34341-2; truncated._ - **[A Novel Data Augmentation Method Based on Denoising Diffusion Probabilistic Model for Fault Diagnosis Under Imbalanced Data](https://space-kitty.com/papers/doi:10.1109/tii.2024.3366991)**. Xiongyan Yang, Tianyi Ye, Xianfeng Yuan et al. · 2024 · IEEE Transactions on Industrial Informatics (doi:10.1109/tii.2024.3366991 · 217 citations). Source: - **[A multimodal comparison of latent denoising diffusion probabilistic models and generative adversarial networks for medical image synthesis](https://space-kitty.com/papers/arxiv:2212.07501)**. Gustav Anton Müller‐Franzes, Jan Moritz Niehues, Firas Khader et al. · 2023 · Scientific Reports (arXiv:2212.07501 · doi:10.1038/s41598-023-39278-0 · 257 citations). Source: > The success of Deep Learning applications critically depends on the quality and scale of the underlying training data. Generative adversarial networks (GANs) can generate arbitrary large datasets, but diversity and fidelity are limited, which has recently been addressed by denoising diffusion probabilistic models (DDPMs) whose superiority has been demonstrated on natural images. In this study, we propose Medfusion, a conditional latent DDPM for medical images. We compare our DDPM-based model against GAN-based models, which constitute the current state-of-the-art in the medical domain. Medfusio… - **[Structured Denoising Diffusion Models in Discrete State-Spaces](https://space-kitty.com/papers/arxiv:2107.03006)**. Jacob Austin, Johnson, Daniel D., Jonathan Ho et al. · 2021 (arXiv:2107.03006 · 303 citations). Source: > Denoising diffusion probabilistic models (DDPMs) (Ho et al. 2020) have shown impressive results on image and waveform generation in continuous state spaces. Here, we introduce Discrete Denoising Diffusion Probabilistic Models (D3PMs), diffusion-like generative models for discrete data that generalize the multinomial diffusion model of Hoogeboom et al. 2021, by going beyond corruption processes with uniform transition probabilities. This includes corruption with transition matrices that mimic Gaussian kernels in continuous space, matrices based on nearest neighbors in embedding space, and matri… - **[Protein structure generation via folding diffusion](https://space-kitty.com/papers/arxiv:2209.15611)**. Kevin Wu, Kevin Yang, Rianne van den Berg et al. · 2024 · Nature Communications (arXiv:2209.15611 · doi:10.1038/s41467-024-45051-2 · 193 citations). Source: > The ability to computationally generate novel yet physically foldable protein structures could lead to new biological discoveries and new treatments targeting yet incurable diseases. Despite recent advances in protein structure prediction, directly generating diverse, novel protein structures from neural networks remains difficult. In this work, we present a new diffusion-based generative model that designs protein backbone structures via a procedure that mirrors the native folding process. We describe protein backbone structure as a series of consecutive angles capturing the relative orientat… - **[Vector Quantized Diffusion Model for Text-to-Image Synthesis](https://space-kitty.com/papers/arxiv:2111.14822)**. Shuyang Gu, Dong Chen, Jianmin Bao et al. · 2021 · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (arXiv:2111.14822 · doi:10.1109/cvpr52688.2022.01043 · 720 citations). Source: > We present the vector quantized diffusion (VQ-Diffusion) model for text-to-image generation. This method is based on a vector quantized variational autoencoder (VQ-VAE) whose latent space is modeled by a conditional variant of the recently developed Denoising Diffusion Probabilistic Model (DDPM). We find that this latent-space method is well-suited for text-to-image generation tasks because it not only eliminates the unidirectional bias with existing methods but also allows us to incorporate a mask-and-replace diffusion strategy to avoid the accumulation of errors, which is a serious problem w… - **[Listen, Denoise, Action! Audio-Driven Motion Synthesis with Diffusion Models](https://space-kitty.com/papers/arxiv:2211.09707)**. Simon Alexanderson, Rajmund Nagy, Jonas Beskow et al. · 2023 · ACM Transactions on Graphics (arXiv:2211.09707 · doi:10.1145/3592458 · 197 citations). Source: > Diffusion models have experienced a surge of interest as highly expressive yet efficiently trainable probabilistic models. We show that these models are an excellent fit for synthesising human motion that co-occurs with audio, e.g., dancing and co-speech gesticulation, since motion is complex and highly ambiguous given audio, calling for a probabilistic description. Specifically, we adapt the DiffWave architecture to model 3D pose sequences, putting Conformers in place of dilated convolutions for improved modelling power. We also demonstrate control over motion style, using classifier-free gui… ## Newest - **[Jailbreak Connectivity: Towards Diverse, Transferable, and Universal Multimodal Large Language Model Jailbreak](https://space-kitty.com/papers/arxiv:2610.38091)**. Xu Zhang, Ziqing Hu, Shuo Han et al. · 2026 (arXiv:2610.38091). Source: - **[StepControl: Long-Horizon Human Motion Generation from Foot-Contact Schedules](https://space-kitty.com/papers/arxiv:2610.38153)**. Hojun Jang, Jinseok Bae, Donggeun Lim et al. · 2026 (arXiv:2610.38153). Source: - **[REEL: Same Frames, Different Answers and Different Costs](https://space-kitty.com/papers/arxiv:2610.38158)**. Xin Liang, Jiangpeng He · 2026 (arXiv:2610.38158). Source: - **[Complementing Text Based Moderation with Multimodal Grounding: Detecting Fraudulent Inducement Behaviors on Short Video Platforms](https://space-kitty.com/papers/arxiv:2610.38185)**. Xiaoyong Yu · 2026 (arXiv:2610.38185). Source: - **[Progressive Coupling Method for Diffusion Posterior Sampling in Ptychography](https://space-kitty.com/papers/arxiv:2610.38207)**. Yihan Qiu, Mingtao Huang, Tianyuan Li et al. · 2026 (arXiv:2610.38207). Source: - **[MEV: A Multi-Event Video Dataset for Long-Take Generation](https://space-kitty.com/papers/arxiv:2610.38241)**. Peiyuan Zhu, Shaoan Xie, Yifan Shen et al. · 2026 (arXiv:2610.38241). Source: - **[NAGO: Noise-Aware Generative Operator via Flow Matching](https://space-kitty.com/papers/arxiv:2610.38273)**. Guanyu Chen, Shengze Xu, Pengwei Liu et al. · 2026 (arXiv:2610.38273). Source: - **[PulmoWorld: An Asynchronous Multimodal Medical World Model for Longitudinal Pneumonia Dynamics](https://space-kitty.com/papers/arxiv:2610.38289)**. Cheng Tang, Xinyu Zhang, Jiawei Chen et al. · 2026 (arXiv:2610.38289). Source: - **[Don't Ask the LLM: Visual Token Pruning with Spectral Structure and Semantic Relevance](https://space-kitty.com/papers/arxiv:2610.38303)**. Intaek Shin, Geuntaek Lim, Sungjune Park et al. · 2026 (arXiv:2610.38303). Source: - **[Aligning Misaligned Benchmarks for Single-View Novel View Synthesis](https://space-kitty.com/papers/arxiv:2610.38349)**. Yang-Che Tseng, Erqun Dong, Jeong-gi Kwak et al. · 2026 (arXiv:2610.38349). Source: --- Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.