研究人员正在探索扩散模型的先进技术，重点是改进图像编辑、数据增强和遗忘能力。新方法旨在通过改进ODE求解器和矢量场平滑来提高图像编辑的稳定性和保真度。对于数据增强，正在开发不确定性指导策略，通过关注信息区域来改进语义分割模型。此外，扩散模型遗忘方面的进展正在取得，研究调查了选择性遗忘和使用稀疏自编码器将概念检测与干预分离，旨在获得更清晰的结果并更好地保留模型质量。
AI
arXiv:2607.01693v1 Announce Type: new Abstract: These notes give a proof-oriented introduction to diffusion models from the viewpoint of sampling, tracing a single arc from classical sampling dynamics to modern diffusion samplers, their error analysis, and inference-time control.…
arXiv:2512.02657v2 Announce Type: replace-cross Abstract: Real-world deployment of text-to-image diffusion models requires continual concept removal as new privacy, copyright, or safety obligations arise over time. Existing unlearning methods, however, are designed for single-ste…
arXiv cs.AI
TIER_1English(EN)·Zhuoxuan Zhang (Yang), Kangqi Ni (Yang), Yuhang Chen (Yang), Mingfu Liang (Yang), Xiaohan Wei (Yang), Yunchen Pu (Yang), Fei Tian (Yang), Chonglin Sun (Yang), Frank Shyu (Yang), Adam (Yang), Song, Sandeep Pandey, Luke Simon, Tianlong Chen, Xi Liu·
arXiv:2607.01170v1 Announce Type: cross Abstract: Generative reasoning re-rankers achieve strong recommendation accuracy by emitting a chain-of-thought before re-ordering a candidate list, but they are slow at inference: an autoregressive (AR) decoder spends one sequential forwar…
arXiv cs.LG
TIER_1English(EN)·Miruna Cretu, John Bradshaw, Patricia Suriana, Saeed Saremi, Omar Mahmood, Kirill Shmilovich, Kangway Chuang, Vishnu Sresht, Colin Grambow·
arXiv:2607.01105v1 Announce Type: new Abstract: We present SynLaD, a latent diffusion framework for small-molecule generation that unifies ligand-based drug design objectives (what to make) with synthetic accessibility (how to make it). Current models typically optimize one objec…
arXiv:2607.00773v1 Announce Type: new Abstract: Discrete diffusion models are widely used for learning and generating discrete distributions. As the generation process is inherently sequential, the acceleration of sampling is of significant importance. In this work, we paralleliz…
arXiv cs.AI
TIER_1English(EN)·Ruikang Zhao, Zhenting Wang, Han Gao, Ligong Han·
arXiv:2607.00208v1 Announce Type: cross Abstract: Reinforcement learning for diffusion large language models (dLLMs) has largely moved to trajectory-aware methods. The current state of the art, TraceRL, holds that random masking is mismatched with the model's inference trajectory…
Generative reasoning re-rankers achieve strong recommendation accuracy by emitting a chain-of-thought before re-ordering a candidate list, but they are slow at inference: an autoregressive (AR) decoder spends one sequential forward pass per reasoning token, and the reasoning trac…
We present SynLaD, a latent diffusion framework for small-molecule generation that unifies ligand-based drug design objectives (what to make) with synthetic accessibility (how to make it). Current models typically optimize one objective at the expense of the other, creating a bot…
Discrete diffusion models are widely used for learning and generating discrete distributions. As the generation process is inherently sequential, the acceleration of sampling is of significant importance. In this work, we parallelize the mainstream $τ$-leaping algorithm for absor…
arXiv cs.LG
TIER_1English(EN)·Onkar Jadhav, Tim French, Matthew Rayson, Nicole L. Jones·
arXiv:2606.31290v1 Announce Type: new Abstract: Diffusion models enable probabilistic super-resolution and conditional generation, but pixel-space methods are computationally expensive and learned latent spaces often lack interpretable uncertainty quantification. We introduce Pat…
arXiv:2605.16399v2 Announce Type: replace-cross Abstract: The inversion of diffusion models plays a central role in image editing. Algebraically reversible ODE solvers provide an appealing approach to diffusion inversion for text-guided image editing, by eliminating the inversion…
arXiv:2602.15008v2 Announce Type: replace Abstract: Diffusion models over discrete spaces have recently shown striking empirical success, yet their theoretical foundations remain incomplete. In this paper, we study the sampling efficiency of score-based discrete diffusion models …
arXiv cs.AI
TIER_1English(EN)·Nikolai R\"ohrich, Julian Glei{\ss}ner, Ahmed H. A. Ibrahim, Silvan Mertes, Tobias Huber·
arXiv:2606.31603v1 Announce Type: cross Abstract: Semantic segmentation models struggle with data sparsity and rare or visually diverse regions, e.g., dense regions or small objects in aerial or autonomous mobility data. While synthetic augmentation is an appealing solution, dire…
arXiv cs.AI
TIER_1English(EN)·Enrico Cassano, Riccardo Renzulli, Rayyan Ahmed, Marco Grangetto, Stephan Alaniz·
arXiv:2606.31699v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) have recently been proposed as interpretable tools for concept-level manipulation, under the assumption that isolated features can serve as controllable intervention points. In this work, we systematical…
arXiv:2604.14790v2 Announce Type: replace Abstract: Interactive Evolutionary Computation (IEC) provides a powerful framework for optimizing subjective criteria such as human preferences and aesthetics, yet it suffers from a fundamental limitation: in high-dimensional generative r…
arXiv cs.AI
TIER_1English(EN)·Jinseong Park, Mijung Park·
arXiv:2510.17917v2 Announce Type: replace-cross Abstract: Data unlearning aims to remove the influence of specific training samples from a trained model. In fine-tuning methods, data unlearning relies primarily on loss maximization over forget samples, which often leads to qualit…
arXiv cs.LG
TIER_1English(EN)·Kundan Kumar, Shreya Das, Simo S\"arkk\"a·
arXiv:2606.31137v1 Announce Type: new Abstract: This paper proposes a Bayesian filtering-based approach for learning the dynamics of a physical system from partial, noisy measurements. We model the system dynamics using a Lagrangian mechanics formulation. As in Lagrangian neural …
Reinforcement learning for diffusion large language models (dLLMs) has largely moved to trajectory-aware methods. The current state of the art, TraceRL, holds that random masking is mismatched with the model's inference trajectory, and it reconstructs that trajectory during train…
Diffusion models enable probabilistic super-resolution and conditional generation, but pixel-space methods are computationally expensive and learned latent spaces often lack interpretable uncertainty quantification. We introduce Patch-PODiff-ViT, a structured latent diffusion fra…
arXiv:2602.11395v2 Announce Type: replace-cross Abstract: Steering diffusion models toward conditions unseen during training typically requires either retraining with conditional inputs or per-step gradient computations, both of which incur substantial computational overhead. We …
arXiv:2603.18907v2 Announce Type: replace Abstract: We propose a new Neural Galerkin Normalizing Flow framework to approximate the transition probability density function of a diffusion process by solving the corresponding Fokker-Planck equation with an atomic initial distributio…
Recent advances in diffusion models have shown impressive performance in controllable image generation and dense prediction tasks. However, existing approaches typically treat diffusion-based controllable generation and dense prediction as separate tasks, overlooking the potentia…
These notes recapitulate the high level mathematical principles behind different techniques for generative modeling. I show the connections between optimal transport and standard techniques such as Schr{ö}dinger bridge and flow matching.
arXiv:2606.27696v1 Announce Type: cross Abstract: In this paper, we are the first to examine the correlations between class frequency and the multi-scale noise schedule within diffusion models. For score-based generative models, low-density regions often lead to inaccurately esti…
In this paper, we are the first to examine the correlations between class frequency and the multi-scale noise schedule within diffusion models. For score-based generative models, low-density regions often lead to inaccurately estimated scores, thereby compromising the generation …
arXiv:2607.00647v1 Announce Type: new Abstract: Training-free guidance (TFG) steers a pretrained diffusion model toward a desired attribute at inference. To be effective, this guidance must be applied from the earliest, high-noise steps of sampling. Because its objective (a class…
arXiv cs.CV
TIER_1English(EN)·H\'ector Laria, Yiping Han, Julian D. Santamaria, Kai Wang, Bogdan Raducanu, Joost van de Weijer, Alexandra Gomez-Villa·
arXiv:2607.00183v1 Announce Type: new Abstract: Adapting pre-trained text-to-image diffusion models, whether to learn new visual concepts or erase unwanted ones, is routinely evaluated on its intended effects alone. We argue this framing is incomplete. Through sparse autoencoder …
arXiv cs.CV
TIER_1English(EN)·Dain Kim, Jinseo Kim, Sungyong Baik·
arXiv:2403.19645v2 Announce Type: replace Abstract: The rapid advancement of diffusion models has enabled the generation of high-fidelity images from textual prompts, yet achieving precise, disentangled control over specific attributes remains a significant challenge. A fundament…
Text-to-image diffusion models achieve impressive visual quality, yet demographic bias remains a challenge, as neutral prompts consistently produce stereotypical representations across gender and race. Existing approaches remain limited by costly retraining or by inference-time i…
Training-free guidance (TFG) steers a pretrained diffusion model toward a desired attribute at inference. To be effective, this guidance must be applied from the earliest, high-noise steps of sampling. Because its objective (a classifier or energy) is defined on clean images, $ε$…
Sparse autoencoders (SAEs) have recently been proposed as interpretable tools for concept-level manipulation, under the assumption that isolated features can serve as controllable intervention points. In this work, we systematically evaluate this assumption in the context of obje…
Semantic segmentation models struggle with data sparsity and rare or visually diverse regions, e.g., dense regions or small objects in aerial or autonomous mobility data. While synthetic augmentation is an appealing solution, directly generating new labeled data risks misalignmen…
arXiv:2606.28785v1 Announce Type: new Abstract: Benefiting from the strong ability to capture data distributions, diffusion models have become powerful tools for solving image inverse problems. The key is to controllably steer the sampling trajectory toward the measurements while…
arXiv:2606.29801v1 Announce Type: new Abstract: Text-to-image diffusion models remain vulnerable to adversarial prompts that elicit disallowed content, motivating reliable inference-time controls. A popular approach is negative guidance, which subtracts a negative prompt directio…
arXiv:2606.28417v1 Announce Type: new Abstract: Recently, diffusion models have been widely adopted in generative modeling and have served as foundational models for many image generation tasks. To control the generation without costly re-training or fine-tuning, many works seek …
arXiv stat.ML
TIER_1English(EN)·Yu Xie, Ludwig Winkler, Lixin Sun, Sarah Lewis, Adam E. Foster, Jos\'e Jim\'enez Luna, Tim Hempel, Michael Gastegger, Yaoyi Chen, Iryna Zaporozhets, Cecilia Clementi, Christopher M. Bishop, Frank No\'e·
arXiv:2602.16634v2 Announce Type: replace Abstract: The rare-event sampling problem has long been the central limiting factor in molecular dynamics (MD), especially in biomolecular simulation. Recently, diffusion models such as BioEmu have emerged as powerful equilibrium samplers…
arXiv:2606.28448v1 Announce Type: cross Abstract: Diffusion posterior samplers for accelerated MRI can reconstruct accurately yet still disagree on the acquired k-space across samples, placing posterior variability on coefficients the scanner has already measured. We identify thi…
arXiv cs.CV
TIER_1English(EN)·Weimin Bai, Yifei Wang, Wenzheng Chen, He Sun·
arXiv:2407.01014v2 Announce Type: replace Abstract: Diffusion models excel in solving imaging inverse problems due to their ability to model complex image priors. However, their reliance on large, clean datasets for training limits their practical use where clean data is scarce. …
<h2> TL;DR </h2> <ul> <li>Efficient and Training-Free Single-Image Diffusion Models is a new approach to generating images that match the internal structure of a single reference image.</li> <li>The model uses a dataset of patches at different scales to compute the score function…
