First systematic column-level sparsity profiling across seven diffusion workloads reveals element-level sparsity overstates hardware savings by up to 78 points and identifies a three-way taxonomy of concentration vs. dispersion.
hub Mixed citations
DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models
Mixed citation behavior. Most common role is background (50%).
42
Pith papers citing it
Background
50% of classified citations
abstract
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 sampling without guidance, their effectiveness for guided sampling has not been well-tested before. In this work, we demonstrate that previous high-order fast samplers suffer from instability issues, and they even become slower than DDIM when the guidance scale grows large. To further speed up guided sampling, we propose DPM-Solver++, a high-order solver for the guided sampling of DPMs. DPM-Solver++ solves the diffusion ODE with the data prediction model and adopts thresholding methods to keep the solution matches training data distribution. We further propose a multistep variant of DPM-Solver++ to address the instability issue by reducing the effective step size. Experiments show that DPM-Solver++ can generate high-quality samples within only 15 to 20 steps for guided sampling by pixel-space and latent-space DPMs.
hub tools
citation-role summary
background 3
method 3
citation-polarity summary
representative citing papers
Derives a conditional-marginal entropy-rate objective for bridge-aware discretization that yields U-shaped schedules and improves low-NFE sample quality on 2D, CIFAR-10, and protein tasks.
Non-monotonic sampling schedules never improve upon monotonic baselines in diffusion models, with performance gaps ranging from substantial to negligible depending on the denoiser.
TMPO uses Softmax Trajectory Balance to match policy probabilities over multiple trajectories to a Boltzmann reward distribution, improving diversity by 9.1% in diffusion alignment tasks.
Grayscale diffusion model generates two-layer RF passives with sub-pixel resolution from partial S-parameters, achieving low error in surrogate predictions and validated on fabricated filters.
DisCa replaces heuristic feature caching with a lightweight learnable neural predictor compatible with distillation, achieving 11.8× acceleration on video diffusion transformers with preserved generation quality.
DiTs use either a two-stage cross-attention circuit or text-token fusion circuit for spatial relations depending on the text encoder, achieving near-perfect in-domain accuracy but differing out-of-domain robustness.
AudioMoG is a mixture-of-guidance sampling technique that combines CFG and AG signals to outperform single-guidance baselines in text-to-audio generation at equivalent speed.
DiffusionNFT performs online RL for diffusion models on the forward process via flow matching and positive-negative contrasts, delivering up to 25x efficiency gains and rapid benchmark improvements over prior reverse-process methods.
MixGRPO speeds up GRPO for flow-based image generators by restricting SDE sampling and optimization to a sliding window while using ODE elsewhere, cutting training time by up to 71% with better alignment performance.
Diffusion trajectory distillation is reframed as operator merging, yielding an optimal variance-driven merging strategy via Pareto dynamic programming in the linear Gaussian case and unavoidable approximation errors from exponential mixture growth in the nonlinear case.
UniEdit-Flow presents tuning-free Uni-Inv and Uni-Edit methods for inversion and editing in flow models that achieve accurate reconstruction and robust region-preserving edits across generative models.
Absorbing discrete diffusion models the conditional distributions of clean data; reparameterizing yields a time-independent RADD that unifies with AO-ARMs and reaches SOTA perplexity among diffusion models on zero-shot language benchmarks.
Latent Consistency Models enable high-fidelity text-to-image generation in 2-4 steps by directly predicting solutions to the probability flow ODE in latent space, distilled from pre-trained LDMs.
ElasticDiT introduces an elastic DiT architecture with adjustable spatial compression and block depth plus Shift Sparse Block Attention and a distilled VAE to enable a single model to cover multiple fidelity-latency points for high-resolution image generation on mobile devices.
FIS-DiT achieves 2.11-2.41x speedup on video DiT models in few-step regimes with negligible quality loss by exploiting frame-wise sparsity and consistency through a training-free interleaved execution strategy.
Generative models learn rules before memorizing data, creating an innovation window whose width depends on dataset size and rule complexity, observed in both diffusion and autoregressive architectures.
LIVEditor-14B applies a new sparse attention method (ISA) that prunes context and uses query-sharpness routing to cut attention latency ~60% with no loss in editing quality on standard benchmarks.
The lookahead drifting model improves upon the drifting model by sequentially computing multiple drifting terms that incorporate higher-order gradient information, leading to better performance on toy examples and CIFAR10.
JFDL allows pre-trained Consistency Models to perform guided image generation post-hoc by aligning flow distributions, reducing FID scores on CIFAR-10 and ImageNet without needing a teacher model.
ConsistencySolver enables high-quality low-step diffusion previews by adapting general linear multistep methods into a lightweight RL-optimized solver, matching multistep DPM-Solver FID with 47% fewer steps and cutting user interaction time by nearly 50%.
PixelDiT generates images in pixel space with a dual-level transformer and reaches 1.61 FID on ImageNet 256, outperforming prior pixel-space models.
NoiseShift learns a resolution-specific mapping from scheduler noise to conditioning noise via lightweight calibration to restore consistency and improve low-resolution generation quality in models like SD3 and Flux.
Multimodal diffusion model generates discrete gate selections and continuous parameters for quantum circuit compilation, claiming better gate counts and noise resilience than prior methods.
citing papers explorer
-
Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success
OpenVLA-OFT fine-tuning boosts LIBERO success rate from 76.5% to 97.1%, speeds action generation 26x, and outperforms baselines on real bimanual dexterous tasks.
-
DyGRO-VLA: Cross-Task Scaling of Vision-Language-Action Models via Dynamic Grouped Residual Optimization
DyGRO-VLA is a two-stage optimization framework for cross-task scaling of Vision-Language-Action models via dynamic grouped residual optimization in RL.