REVIEW 27 cited by
SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
SiT: Exploring Flow and Diffusion-based Generative Models with Scalable Interpolant Transformers
read the original abstract
We present Scalable Interpolant Transformers (SiT), a family of generative models built on the backbone of Diffusion Transformers (DiT). The interpolant framework, which allows for connecting two distributions in a more flexible way than standard diffusion models, makes possible a modular study of various design choices impacting generative models built on dynamical transport: learning in discrete or continuous time, the objective function, the interpolant that connects the distributions, and deterministic or stochastic sampling. By carefully introducing the above ingredients, SiT surpasses DiT uniformly across model sizes on the conditional ImageNet 256x256 and 512x512 benchmark using the exact same model structure, number of parameters, and GFLOPs. By exploring various diffusion coefficients, which can be tuned separately from learning, SiT achieves an FID-50K score of 2.06 and 2.62, respectively.
Forward citations
Cited by 27 Pith papers
-
What Time Is It? How Data Geometry Makes Time Conditioning Optional for Flow Matching
Data geometry makes time identifiable from noisy interpolants at rate O(1/sqrt(d-k)), rendering the time-blindness gap asymptotically negligible relative to coupling variance.
-
Balancing Image Compression and Generation with Bootstrapped Tokenization
SelfBootTok decomposes image tokens into global and local groups via self-bootstrapped learning, enabling generators to use only global tokens for ~40% less computation and a new SOTA gFID of 1.56 with 64 tokens.
-
Posterior Augmented Flow Matching
PAFM augments flow matching with an importance-sampled mixture over an approximate posterior of target completions, yielding an unbiased lower-variance estimator that improves FID by up to 3.4 on ImageNet and CC12M.
-
GVCC: Zero-Shot Video Compression via Codebook-Driven Stochastic Rectified Flow
GVCC achieves the lowest LPIPS on UVG at bitrates down to 0.003 bpp by encoding stochastic innovations in a marginal-preserving stochastic process derived from a pretrained rectified-flow video model, with 65% LPIPS r...
-
Setting-Matched and Semantics-Scaled Benchmarking of One-Step Generative Models Against Multistep Diffusion and Flow Models
Matched benchmarking reveals FID misleads in few-step regimes under CFG, prompting CLIP-scaled and PickScore-scaled FID and IS variants for better semantic evaluation of one-step image generators.
-
PixelU: A U-Shaped Transformer for Efficient End-to-End Pixel Diffusion
PixelU is a minimalist U-shaped Diffusion Transformer for pixel-space diffusion that decouples frequencies with zero-cost skip connections and constant-channel downsampling, outperforming baselines like JiT-G at 1/3 t...
-
MIMFlow: Integrating Masked Image Modeling with Normalizing Flows for End-to-End Image Generation
MIMFlow uses a VAE on masked images to feed semantic latents to a normalizing flow while a decoder handles high-frequency details, reporting FID 2.50 and 71.3% linear probing on ImageNet 256x256 with 128 tokens.
-
MIMFlow: Integrating Masked Image Modeling with Normalizing Flows for End-to-End Image Generation
End-to-end masked-image VAE plus normalizing flow yields FID 2.50 on ImageNet 256 with 128 tokens and higher linear-probe accuracy than unmasked counterparts.
-
MIMFlow: Integrating Masked Image Modeling with Normalizing Flows for End-to-End Image Generation
MIMFlow is an end-to-end model that routes semantic latents through a normalizing flow while a decoder handles high-frequency pixels, reporting FID 2.50 and 71.3% linear probing accuracy on ImageNet 256x256 with 128 tokens.
-
IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder
IDEAL improves discrete representation autoencoders by jointly aligning quantized tokens with shallow and deep VFM features, reporting 0.61 rFID on ImageNet and 1.89 gFID for autoregressive image generation.
-
DSA: Dynamic Step Allocation for Fast Autoregressive Video Generation
DSA adds a jointly trained confidence head to autoregressive video diffusion models that dynamically allocates fewer or more denoising steps per frame, achieving 22.63 FPS real-time generation on H100 while matching V...
-
SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation
SRC-Flow compresses RAE features via a Semantic Representation Compressor into a low-dimensional space, enabling normalizing flows to reach gFID 1.65 on ImageNet 256x256 and 2.07 on 512x512 while retaining exact likelihoods.
-
SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation
SRC-Flow compresses RAE features into a low-dimensional semantic space with a Semantic Representation Compressor, enabling normalizing flows to achieve SOTA gFID scores of 1.65 and 2.07 on ImageNet 256x256 and 512x512...
-
The two clocks and the innovation window: When and how generative models learn rules
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.
-
Denoising, Fast and Slow: Difficulty-Aware Adaptive Sampling for Image Generation
Patch Forcing enables diffusion models to denoise image patches at varying rates based on predicted difficulty, advancing easier regions first to improve context and achieve better generation quality on ImageNet while...
-
PixelGen: Improving Pixel Diffusion with Perceptual Supervision
PixelGen augments pixel diffusion with gated perceptual supervision to reach FID 5.11 on ImageNet-256 and GenEval 0.79 in text-to-image, narrowing the gap to latent methods without VAEs.
-
DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation
DeCo decouples high- and low-frequency generation in pixel diffusion via a DiT plus lightweight decoder and a frequency-aware flow-matching loss, reaching FID 1.62 at 256x256 and 2.22 at 512x512 on ImageNet while clos...
-
Flow marching for a generative PDE foundation model
Flow Marching jointly samples noise and physical time to learn a velocity field for generative PDE modeling, paired with a latent autoencoder and efficient transformer for large-scale pretraining on 2.5M trajectories.
-
Cross-Distribution Diffusion Priors-Driven Iterative Reconstruction for Sparse-View CT
CDPIR integrates cross-distribution diffusion priors from a Scalable Interpolant Transformer trained with classifier-free guidance into model-based iterative reconstruction to improve sparse-view CT under out-of-distr...
-
Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps
Diffusion models improve generation quality via inference-time search over noise candidates guided by verifiers and algorithms, yielding gains beyond denoising step scaling on class- and text-conditioned benchmarks.
-
EventFlow: Forecasting Temporal Point Processes with Flow Matching
EventFlow applies flow matching to learn joint distributions over event times for temporal point processes, reporting 20-53% lower forecast error than autoregressive baselines on standard TPP benchmarks with fewer sam...
-
HistoFID- Calibrating Frechet-distance evaluation across pathology foundation models
Raw Fréchet distances in pathology vary ~30-fold across encoders; dividing by each encoder's own within-cohort floor cuts cross-encoder variation by ~89% within and ~58% across cohorts.
-
CaloTrilogy: Toward a Breakthrough in One-Step, End-to-End, Physics-Guided Shower Generation for Modern Calorimeters
Presents CaloTrilogy, a unified one-step generative model for high-granularity calorimeter showers that combines velocity field integration, learned priors, and physics losses to match SOTA quality.
-
FrequencyBooster: Full-Frequency Modeling for High-Fidelity Pixel Diffusion
FrequencyBooster reports state-of-the-art FID scores of 1.60 at 256x256 and 1.69 at 512x512 for pixel diffusion by using a specialized decoder for full-frequency modeling.
-
Discrete Meanflow Training Curriculum
A DMF curriculum initialized from pretrained flow models achieves one-step FID 3.36 on CIFAR-10 after only 2000 epochs by exploiting a discretized consistency property in the Meanflow objective.
-
Seedream 3.0 Technical Report
Seedream 3.0 improves bilingual image generation through doubled defect-aware data, mixed-resolution training, cross-modality RoPE, representation alignment, aesthetic SFT, VLM reward modeling, and importance-aware ti...
-
Flow Matching Guide and Code
Flow Matching is a generative modeling framework with mathematical foundations, design choices, extensions, and open-source PyTorch code for applications like image and text generation.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.