Picture for Jonas Kohler

Jonas Kohler

FlexiDiT: Your Diffusion Transformer Can Easily Generate High-Quality Samples with Less Compute

Add code
Feb 27, 2025
Viaarxiv icon

Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment

Add code
Jan 31, 2025
Figure 1 for Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment
Figure 2 for Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment
Figure 3 for Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment
Figure 4 for Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment
Viaarxiv icon

Movie Gen: A Cast of Media Foundation Models

Add code
Oct 17, 2024
Figure 1 for Movie Gen: A Cast of Media Foundation Models
Figure 2 for Movie Gen: A Cast of Media Foundation Models
Figure 3 for Movie Gen: A Cast of Media Foundation Models
Figure 4 for Movie Gen: A Cast of Media Foundation Models
Viaarxiv icon

Imagine Flash: Accelerating Emu Diffusion Models with Backward Distillation

Add code
May 08, 2024
Viaarxiv icon

fMPI: Fast Novel View Synthesis in the Wild with Layered Scene Representations

Add code
Dec 26, 2023
Viaarxiv icon

Adaptive Guidance: Training-free Acceleration of Conditional Diffusion Models

Add code
Dec 19, 2023
Viaarxiv icon

Cache Me if You Can: Accelerating Diffusion Models through Block Caching

Add code
Dec 06, 2023
Figure 1 for Cache Me if You Can: Accelerating Diffusion Models through Block Caching
Figure 2 for Cache Me if You Can: Accelerating Diffusion Models through Block Caching
Figure 3 for Cache Me if You Can: Accelerating Diffusion Models through Block Caching
Figure 4 for Cache Me if You Can: Accelerating Diffusion Models through Block Caching
Viaarxiv icon

Synthesizing Speech from Intracranial Depth Electrodes using an Encoder-Decoder Framework

Add code
Nov 02, 2021
Figure 1 for Synthesizing Speech from Intracranial Depth Electrodes using an Encoder-Decoder Framework
Figure 2 for Synthesizing Speech from Intracranial Depth Electrodes using an Encoder-Decoder Framework
Viaarxiv icon

Safe Deep Reinforcement Learning for Multi-Agent Systems with Continuous Action Spaces

Add code
Aug 11, 2021
Figure 1 for Safe Deep Reinforcement Learning for Multi-Agent Systems with Continuous Action Spaces
Figure 2 for Safe Deep Reinforcement Learning for Multi-Agent Systems with Continuous Action Spaces
Figure 3 for Safe Deep Reinforcement Learning for Multi-Agent Systems with Continuous Action Spaces
Figure 4 for Safe Deep Reinforcement Learning for Multi-Agent Systems with Continuous Action Spaces
Viaarxiv icon

Vanishing Curvature and the Power of Adaptive Methods in Randomly Initialized Deep Networks

Add code
Jun 07, 2021
Figure 1 for Vanishing Curvature and the Power of Adaptive Methods in Randomly Initialized Deep Networks
Figure 2 for Vanishing Curvature and the Power of Adaptive Methods in Randomly Initialized Deep Networks
Figure 3 for Vanishing Curvature and the Power of Adaptive Methods in Randomly Initialized Deep Networks
Figure 4 for Vanishing Curvature and the Power of Adaptive Methods in Randomly Initialized Deep Networks
Viaarxiv icon