13 papers
Latent Space Jam: Layout Guided Video Synthesis
Asaf Joseph, Moshe Mandel, Ohad Fried
Adding control to video generation often requires training, which demands additional computational resources, data, or architectural changes. We propose a training-free approach th…
SoftCache: Adaptive Step Size Caching for Diffusion Transformers
Shimon Malnick, Nadav Orenstein, Ohad Fried +1
Diffusion and flow models generate a sample over iterative steps, each a full pass through a large network, and that pass dominates generation cost. Training-free caching accelerat…
Tracing Generated Samples to Training-Data Clusters in Flow-Matching Models
Rania Briq, Ohad Fried, Michael Kamp +1
Understanding which training samples influence a generated image is an important problem in generative modeling. In flow matching, training samples influence the generated image th…
Gefen: Optimized Stochastic Optimizer
Nadav Benedek, Tomer Koren, Ohad Fried
AdamW is a default optimizer for deep learning, but its moment states add two parameter-sized buffers to training memory, increasing the cost of large-scale pretraining. We propose…
NIV: Neural Axis Variations for Variable Font Generation
Nadav Benedek, Ariel Shamir, Ohad Fried
Variable fonts enable continuous variation of glyph geometry along semantic design axes such as weight, width, slant, and optical size. However, constructing a variable font from a…
Optimal Transport Flow Matching by Design
Shimon Malnick, Matan Rusanovsky, Ohad Fried +1
Flow matching models learn to transport samples from a simple prior distribution to a complex data distribution. When prior-data pairs are coupled via optimal transport (OT), the l…
Exploring and Exploiting Stability in Latent Flow Matching
Rania Briq, Michael Kamp, Ohad Fried +2
In this work, we show that Latent Flow-Matching (LFM) models are robust to different types of perturbations, including data reduction and model capacity shrinkage. We characterize…
The Amazing Stability of Flow Matching
Rania Briq, Michael Kamp, Ohad Fried +2
The success of deep generative models in generating high-quality and diverse samples is often attributed to particular architectures and large training datasets. In this paper, we…
Copy-Trasform-Paste: Zero-Shot Object-Object Alignment Guided by Vision-Language and Geometric Constraints
Rotem Gatenyo, Ohad Fried
We study zero-shot 3D alignment of two given meshes, using a text prompt describing their spatial relation -- an essential capability for content creation and scene assembly. Earli…
Express4D: Expressive, Friendly, and Extensible 4D Facial Motion Generation Benchmark
Yaron Aloni, Rotem Shalev-Arkushin, Yonatan Shafir +3
Dynamic facial expression generation from natural language is a crucial task in Computer Graphics, with applications in Animation, Virtual Avatars, and Human-Computer Interaction.…
REED-VAE: RE-Encode Decode Training for Iterative Image Editing with Diffusion Models
Gal Almog, Ariel Shamir, Ohad Fried
While latent diffusion models achieve impressive image editing results, their application to iterative editing of the same image is severely restricted. When trying to apply consec…
Tiled Diffusion
Or Madar, Ohad Fried
Image tiling -- the seamless connection of disparate images to create a coherent visual field -- is crucial for applications such as texture creation, video game asset development,…
Memories of Forgotten Concepts
Matan Rusanovsky, Shimon Malnick, Amir Jevnisek +2
Diffusion models dominate the space of text-to-image generation, yet they may produce undesirable outputs, including explicit content or private data. To mitigate this, concept abl…