Image, video and audio generation — diffusion models, flow matching and the systems built on top of them.
arXiv:2605.06272v2 Announce Type: replace Abstract: While generative modeling has achieved remarkable success on tasks like natural language-conditioned image generation, enabling model adaptation fr...
arXiv:2610.00414v1 Announce Type: new Abstract: Foundation models pretrained on large-scale datasets demonstrate strong transferability to medical imaging tasks. However, understanding how their late...
arXiv:2610.01012v1 Announce Type: new Abstract: Video-to-speech synthesis aims to generate natural-sounding speech from silent talking-face videos while ensuring phonetic accuracy. A fundamental chal...
arXiv:2610.01092v1 Announce Type: new Abstract: Video generation models are increasingly being explored as world simulators for embodied planning and learning. To do so effectively, these models must...
arXiv:2610.01162v1 Announce Type: new Abstract: Reliable video world models could provide scalable predictive environments for robot learning, planning, and evaluation. However, generated robot video...
arXiv:2610.01233v1 Announce Type: new Abstract: Flow matching is central to 3D generation, yet in practice its reinforcement learning (RL) methods are largely adapted from 2D visual generation. Repre...
arXiv:2610.01681v1 Announce Type: new Abstract: Unified models are trained for both instruction-based image editing and text-to-image (T2I) generation, but standard editing pipelines keep source-imag...
arXiv:2610.01969v1 Announce Type: new Abstract: Concept erasure aims to remove a target concept, such as a copyrighted style, a recognizable character, or unsafe content, from a pretrained text-to-im...
arXiv:2610.02044v1 Announce Type: new Abstract: Recent advances in rectified flow-based image-to-3D generative models have enabled high-fidelity 3D asset generation. Building on this, a growing line...
arXiv:2610.02153v1 Announce Type: new Abstract: Long-horizon autoregressive video generation is limited by a finite context window. When an object or scene falls out of context, its fine-grained visu...
arXiv:2610.02208v1 Announce Type: new Abstract: Sphere Encoder is an autoencoder that generates images by decoding random points from a high-dimensional latent sphere. We identify two limitations of...
arXiv:2610.00895v1 Announce Type: cross Abstract: Foundation models remain vulnerable to spurious correlations and ``Clever Hans'' strategies. Explainable machine learning can find and remove such st...
arXiv:2605.11506v2 Announce Type: replace Abstract: Score-based diffusion models achieve state-of-the-art performance for inverse problems, but their practical deployment is hindered by long inferenc...
arXiv:2605.21244v2 Announce Type: replace Abstract: Super-Resolution (SR) has advanced rapidly in recent years, with diffusion-based models achieving unprecedented fidelity at the cost of introducing...
arXiv:2606.08091v2 Announce Type: replace Abstract: Agentic long video generation requires planning, tool orchestration, and cross-clip coordination over a long horizon. Most existing video agents ei...
arXiv:2607.08020v2 Announce Type: replace Abstract: Autoregressive video diffusion enables efficient streaming and long-horizon video generation, but repeatedly reusing generated latents as causal co...
arXiv:2610.01933v1 Announce Type: new Abstract: Inference-time scaling adapts pretrained diffusion models to new sampling tasks without additional training. Existing methods rely primarily on Monte C...
arXiv:2610.01522v1 Announce Type: cross Abstract: Many scientific and machine learning systems, from molecular dynamics to diffusion models and beyond, are governed by stochastic dynamics with low-di...
arXiv:2610.02128v1 Announce Type: cross Abstract: Many applications in statistics, economics, and physics require sampling from high-dimensional categorical distributions with local dependence struct...
STAGE is a training‑free, closed‑form framework for concept erasure in native text‑to‑3D generators. It treats erasure as a stage‑aware editing problem, applying low‑dimensional affine corrections separately to the structural and appearance stages of the pipeline. Experiments on the TRELLIS generator show that STAGE outperforms adapted baselines, achieving a composite score of 66.7 versus 53.2 across 15 shape, material, and object concepts.