LLaDA is a scalable diffusion-based language model that matches autoregressive LLMs like LLaMA3 8B on tasks and surpasses GPT-4o on reversal poem completion.
hub Canonical reference
Scalable diffusion models with transformers
Canonical reference. 75% of citing Pith papers cite this work as background.
15
Pith papers citing it
Background
75% of classified citations
hub tools
citation-role summary
background 7
method 1
citation-polarity summary
representative citing papers
ICEdit achieves state-of-the-art instructional image editing in Diffusion Transformers via in-context generation, requiring only 0.1% of prior training data and 1% trainable parameters.
RAE v2 reaches gFID 1.06 on ImageNet-256 in 80 epochs by combining multi-layer encoder sums, complementary REPA targets, and free guidance via output reparameterization.
Sec2Drum-DAC renders drum audio from symbolic inputs via diffusion on PCA-reduced DAC latents, improving spectral and transient metrics over regression baselines on 1733 held-out windows.
Sparse RFNNs with sSVD via Lanczos-Golub-Kahan bidiagonalization maintain accuracy while improving efficiency and robustness for 1D steady convection-diffusion equations with strong advection.
A recursive sparse MoE framework integrated into diffusion models iteratively refines visual tokens via gated module selection to improve structured reasoning and image generation performance.
CoreFlow is a low-rank matrix generative model that trains normalizing flows on shared subspaces to improve efficiency and quality for high-dimensional limited-sample data, including incomplete matrices.
VideoPhy benchmark shows state-of-the-art text-to-video models follow physical commonsense and text prompts in only 39.6% of cases for the best model.
PipeFusion applies patch partitioning and pipeline parallelism with one-step stale feature reuse to reduce communication overhead in DiT inference, reporting SOTA results on 8x L40 GPUs for Pixart, SD3, and Flux.1.
MONET is an open 104.9M image-text pair dataset created via safety filtering, deduplication, and multi-VLM recaptioning from 2.9B raw pairs, validated by training a competitive 4B-parameter latent diffusion model.
Using understanding tasks as direct supervision during post-training improves image generation and editing in unified multimodal models.
Aligning the DDIM forward diffusion process with flow-matching manifold evolution enables high-quality generation without time conditioning, and class-conditional synthesis is possible with an unconditional denoiser by using separate time spaces per class.
LTX-2 generates high-quality synchronized audiovisual content from text prompts via an asymmetric 14B-video / 5B-audio dual-stream transformer with cross-attention and modality-aware guidance.
The survey frames VLA models as pipelines that generate progressively grounded action tokens and classifies those tokens into eight types to guide future development.
citing papers explorer
-
Large Language Diffusion Models
LLaDA is a scalable diffusion-based language model that matches autoregressive LLMs like LLaMA3 8B on tasks and surpasses GPT-4o on reversal poem completion.
-
In-Context Edit: Enabling Instructional Image Editing with In-Context Generation in Large Scale Diffusion Transformer
ICEdit achieves state-of-the-art instructional image editing in Diffusion Transformers via in-context generation, requiring only 0.1% of prior training data and 1% trainable parameters.
-
Improved Baselines with Representation Autoencoders
RAE v2 reaches gFID 1.06 on ImageNet-256 in 80 epochs by combining multi-layer encoder sums, complementary REPA targets, and free guidance via output reparameterization.
-
Seconds-Aligned PCA-DAC Latent Diffusion for Symbolic-to-Audio Drum Rendering
Sec2Drum-DAC renders drum audio from symbolic inputs via diffusion on PCA-reduced DAC latents, improving spectral and transient metrics over regression baselines on 1733 held-out windows.
-
Sparse Random-Feature Neural Networks with Krylov-Based SVD for Singularly Perturbed ODE
Sparse RFNNs with sSVD via Lanczos-Golub-Kahan bidiagonalization maintain accuracy while improving efficiency and robustness for 1D steady convection-diffusion equations with strong advection.
-
The Thinking Pixel: Recursive Sparse Reasoning in Multimodal Diffusion Latents
A recursive sparse MoE framework integrated into diffusion models iteratively refines visual tokens via gated module selection to improve structured reasoning and image generation performance.
-
CoreFlow: Low-Rank Matrix Generative Models
CoreFlow is a low-rank matrix generative model that trains normalizing flows on shared subspaces to improve efficiency and quality for high-dimensional limited-sample data, including incomplete matrices.
-
VideoPhy: Evaluating Physical Commonsense for Video Generation
VideoPhy benchmark shows state-of-the-art text-to-video models follow physical commonsense and text prompts in only 39.6% of cases for the best model.
-
PipeFusion: Patch-level Pipeline Parallelism for Diffusion Transformers Inference
PipeFusion applies patch partitioning and pipeline parallelism with one-step stale feature reuse to reduce communication overhead in DiT inference, reporting SOTA results on 8x L40 GPUs for Pixart, SD3, and Flux.1.
-
MONET: A Massive, Open, Non-redundant and Enriched Text-to-image dataset
MONET is an open 104.9M image-text pair dataset created via safety filtering, deduplication, and multi-VLM recaptioning from 2.9B raw pairs, validated by training a competitive 4B-parameter latent diffusion model.
-
Steering Visual Generation in Unified Multimodal Models with Understanding Supervision
Using understanding tasks as direct supervision during post-training improves image generation and editing in unified multimodal models.
-
Exploring Time Conditioning in Diffusion Generative Models from Disjoint Noisy Data Manifolds
Aligning the DDIM forward diffusion process with flow-matching manifold evolution enables high-quality generation without time conditioning, and class-conditional synthesis is possible with an unconditional denoiser by using separate time spaces per class.
-
LTX-2: Efficient Joint Audio-Visual Foundation Model
LTX-2 generates high-quality synchronized audiovisual content from text prompts via an asymmetric 14B-video / 5B-audio dual-stream transformer with cross-attention and modality-aware guidance.
-
A Survey on Vision-Language-Action Models: An Action Tokenization Perspective
The survey frames VLA models as pipelines that generate progressively grounded action tokens and classifies those tokens into eight types to guide future development.
- Beyond Text Prompts: Visual-to-Visual Generation as A Unified Paradigm