GEAR jointly trains VQ tokenizer and AR generator end-to-end via dual hard/soft read-out and representation alignment, achieving up to 10x faster ImageNet gFID convergence than LlamaGen-REPA while generalizing across quantizers and to text-to-image.
hub Mixed citations
Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
Mixed citation behavior. Most common role is background (43%).
91
Pith papers citing it
Background
43% of classified citations
abstract
Recent studies have shown that the denoising process in (generative) diffusion models can induce meaningful (discriminative) representations inside the model, though the quality of these representations still lags behind those learned through recent self-supervised learning methods. We argue that one main bottleneck in training large-scale diffusion models for generation lies in effectively learning these representations. Moreover, training can be made easier by incorporating high-quality external visual representations, rather than relying solely on the diffusion models to learn them independently. We study this by introducing a straightforward regularization called REPresentation Alignment (REPA), which aligns the projections of noisy input hidden states in denoising networks with clean image representations obtained from external, pretrained visual encoders. The results are striking: our simple strategy yields significant improvements in both training efficiency and generation quality when applied to popular diffusion and flow-based transformers, such as DiTs and SiTs. For instance, our method can speed up SiT training by over 17.5$\times$, matching the performance (without classifier-free guidance) of a SiT-XL model trained for 7M steps in less than 400K steps. In terms of final generation quality, our approach achieves state-of-the-art results of FID=1.42 using classifier-free guidance with the guidance interval.
hub tools
citation-role summary
background 10
baseline 6
method 4
extension 1
citation-polarity summary
representative citing papers
PRA approximates sequential rollout training in parallel for pixel-space AR models via intermediate states and a pixel decoder, achieving FID 2.58 (135M params) and 1.94 (511M params) on ImageNet-1K 256x256, new SOTA among pixel-space AR models.
Continuous language diffusion works by entering high-margin decoder basins where frozen T5 embeddings recover 93-96% of native decisions and linear readouts reach 97.9% agreement, implying models should be evaluated as representation-decoder systems.
MaskAlign uses random token-subset alignment and pre-mask mixing to reduce diffusion models' reliance on complete clean-image token sets during representation alignment.
CoFi-UCGen achieves both coarse- and fine-grained unsupervised conditional image generation by using bit-codes for structured latent space and hierarchical modulation in diffusion models.
VPE inserts an internal autoregressive visual semantic token generation step to guide image token production in unified models, reporting faster convergence, higher quality, and superior editing preservation (PSNR 26.76 vs 19.92) versus external alternatives.
A large-scale empirical study across tokenizers and diffusion backbones identifies Velocity Irreducible Variance (VIV) as one of the most stable predictors of latent diffusion generation quality.
Neural reconstruction losses in VAEs reduce latent information content and produce more isotropic latent geometries with even uncertainty distribution.
STROP learns variable-length discrete visual programs for images by training a length head against frozen DINOv3 features in a four-phase curriculum while bypassing pixel reconstruction.
Uni-Edit introduces a data synthesis pipeline turning VQA data into reasoning-intensive editing instructions, enabling single-task tuning that boosts all three capabilities in models like BAGEL and Janus-Pro.
Optimal INR freeze depth matches highest weight stable rank layer; SAEs reveal SIREN atoms are localized while FFMLP atoms trace cohort contours with causal impact on PSNR.
Prologue adds a small set of learnable tokens trained exclusively with AR cross-entropy loss to decouple generation from reconstruction in autoregressive visual models, yielding lower gFID on ImageNet 256x256.
M²-REPA decouples modality-specific features from diffusion intermediates and aligns them to complementary expert foundation models via a multi-modal alignment loss and modality-specific decoupling regularization for improved multimodal video generation.
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.
A 3D-grounded autoencoder and diffusion transformer allow direct generation of 3D scenes in an implicit latent space using a fixed 1K-token representation for arbitrary views and resolutions.
3D-Fixer performs in-place 3D asset completion from single-view partial point clouds via coarse-to-fine generation with ORFA conditioning, plus a new ARSG-110K dataset, to achieve higher geometric accuracy than MIDI and Gen3DSR while keeping diffusion efficiency.
LatentTSF improves time series forecasting accuracy and representation quality by shifting prediction from observation space to a learned latent state space via autoencoding.
CCDD defines a joint multimodal diffusion on continuous representation space and discrete token space to combine expressivity with explicit token supervision for diffusion language models.
Distribution-wise rewards with subset-replace strategy and post-hoc merging improve FID-50K on SiT (8.30 to 5.77) and EDM2 (3.74 to 3.52) while preserving diversity.
TerraDiT-Ω generates satellite imagery from native geospatial primitives via Geometry-Aware Local Attention and outperforms dense and sparse control baselines while boosting downstream GeoAI tasks.
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.
Proposes LHE, SRB, and AFL components in a semantics-first latent framework that yields better 3D MRI reconstruction and cross-contrast synthesis on two public datasets.
GAM splits a geometric foundation model to enable language-conditioned future geometry prediction and action decoding for robot policies, claiming superior performance on manipulation benchmarks.
HYDRA-X presents the first unified multimodal model using a single ViT for holistic image-video tokenization, with ablations on attention and compression plus a latent-level editing improvement.
citing papers explorer
-
GEAR: Guided End-to-End AutoRegression for Image Synthesis
GEAR jointly trains VQ tokenizer and AR generator end-to-end via dual hard/soft read-out and representation alignment, achieving up to 10x faster ImageNet gFID convergence than LlamaGen-REPA while generalizing across quantizers and to text-to-image.
-
Parallel Rollout Approximation for Pixel-Space Autoregressive Image Generation
PRA approximates sequential rollout training in parallel for pixel-space AR models via intermediate states and a pixel decoder, achieving FID 2.58 (135M params) and 1.94 (511M params) on ImageNet-1K 256x256, new SOTA among pixel-space AR models.
-
Continuous Language Diffusion as a Decoder-Interface Problem
Continuous language diffusion works by entering high-margin decoder basins where frozen T5 embeddings recover 93-96% of native decisions and linear readouts reach 97.9% agreement, implying models should be evaluated as representation-decoder systems.
-
MaskAlign: Token-Subset Representation Alignment for Efficient Diffusion Training
MaskAlign uses random token-subset alignment and pre-mask mixing to reduce diffusion models' reliance on complete clean-image token sets during representation alignment.
-
CoFi-UCGen: Coarse-to-Fine Unsupervised Conditional Generation without Label Priors
CoFi-UCGen achieves both coarse- and fine-grained unsupervised conditional image generation by using bit-codes for structured latent space and hierarchical modulation in diffusion models.
-
Imagine Before You Draw: Visual Prompt Engineering for Image Generation
VPE inserts an internal autoregressive visual semantic token generation step to guide image token production in unified models, reporting faster convergence, higher quality, and superior editing preservation (PSNR 26.76 vs 19.92) versus external alternatives.
-
Diffusing in the Right Space: A Systematic Study of Latent Diffusability
A large-scale empirical study across tokenizers and diffusion backbones identifies Velocity Irreducible Variance (VIV) as one of the most stable predictors of latent diffusion generation quality.
-
How Neural Losses Shape VAE Latents
Neural reconstruction losses in VAEs reduce latent information content and produce more isotropic latent geometries with even uncertainty distribution.
-
Structure over Pixels: Learning Variable-Length Visual Programs
STROP learns variable-length discrete visual programs for images by training a length head against frozen DINOv3 features in a four-phase curriculum while bypassing pixel reconstruction.
-
Uni-Edit: Intelligent Editing Is A General Task For Unified Model Tuning
Uni-Edit introduces a data synthesis pipeline turning VQA data into reasoning-intensive editing instructions, enabling single-task tuning that boosts all three capabilities in models like BAGEL and Janus-Pro.
-
What Cohort INRs Encode and Where to Freeze Them
Optimal INR freeze depth matches highest weight stable rank layer; SAEs reveal SIREN atoms are localized while FFMLP atoms trace cohort contours with causal impact on PSNR.
-
Autoregressive Visual Generation Needs a Prologue
Prologue adds a small set of learnable tokens trained exclusively with AR cross-entropy loss to decouple generation from reconstruction in autoregressive visual models, yielding lower gFID on ImageNet 256x256.
-
Divide and Conquer: Decoupled Representation Alignment for Multimodal World Models
M²-REPA decouples modality-specific features from diffusion intermediates and aligns them to complementary expert foundation models via a multi-modal alignment loss and modality-specific decoupling regularization for improved multimodal video generation.
-
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.
-
Any 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale
A 3D-grounded autoencoder and diffusion transformer allow direct generation of 3D scenes in an implicit latent space using a fixed 1K-token representation for arbitrary views and resolutions.
-
3D-Fixer: Coarse-to-Fine In-place Completion for 3D Scenes from a Single Image
3D-Fixer performs in-place 3D asset completion from single-view partial point clouds via coarse-to-fine generation with ORFA conditioning, plus a new ARSG-110K dataset, to achieve higher geometric accuracy than MIDI and Gen3DSR while keeping diffusion efficiency.
-
From Observations to States: Latent Time Series Forecasting
LatentTSF improves time series forecasting accuracy and representation quality by shifting prediction from observation space to a learned latent state space via autoencoding.
-
Coevolutionary Continuous Discrete Diffusion: Make Your Diffusion Language Model a Latent Reasoner
CCDD defines a joint multimodal diffusion on continuous representation space and discrete token space to combine expressivity with explicit token supervision for diffusion language models.
-
Optimizing Visual Generative Models via Distribution-wise Rewards
Distribution-wise rewards with subset-replace strategy and post-hoc merging improve FID-50K on SiT (8.30 to 5.77) and EDM2 (3.74 to 3.52) while preserving diversity.
-
TerraDiT-$\Omega$: Unified Spatial Control for Satellite Image Synthesis with Any Geospatial Primitive
TerraDiT-Ω generates satellite imagery from native geospatial primitives via Geometry-Aware Local Attention and outperforms dense and sparse control baselines while boosting downstream GeoAI tasks.
-
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.
-
Recover Semantics First, Generate Better: Improved Latent Modeling for 3D MRI Reconstruction and Cross-Contrast Synthesis
Proposes LHE, SRB, and AFL components in a semantics-first latent framework that yields better 3D MRI reconstruction and cross-contrast synthesis on two public datasets.
-
Geometric Action Model for Robot Policy Learning
GAM splits a geometric foundation model to enable language-conditioned future geometry prediction and action decoding for robot policies, claiming superior performance on manipulation benchmarks.
-
HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers
HYDRA-X presents the first unified multimodal model using a single ViT for holistic image-video tokenization, with ablations on attention and compression plus a latent-level editing improvement.
-
LARA: Latent Action Representation Alignment for Vision-Language-Action Models
LARA jointly optimizes LAM and VLA models via representation alignment to improve robotic manipulation performance using human videos.
-
Beyond Skeletons: Learning Animation Directly from Driving Videos with Same2X Training Strategy
DirectAnimator bypasses pose extraction using a Driving Cue Triplet and Same2X training strategy to achieve state-of-the-art human animation quality and robustness from raw videos.
-
Right Makes Might: Aligning Verified Hidden States Empowers RL Reasoning
Hidden-Align adds an auxiliary loss to align hidden states of correct reasoning paths at the pre-answer token in RLVR, improving pass@1 by 3.8-6.2 points over DAPO on eight math benchmarks for Qwen3 models of 1.7B-14B scale.
-
Geometry-Aware Implicit Memory for Video World Models
GIM-World adds a camera-queryable geometry distillation head and pruning rule to implicit memory in video world models, claiming better long-horizon geometric consistency on the MIND benchmark than explicit and implicit baselines.
-
Generate in Reconstruction Space, Match in Semantic Space: Transport Geometry for One-Step Generation
Matching in semantic SSL feature space via Sinkhorn divergence enables effective one-step generation on ImageNet by inducing compact geometry for distribution matching, with training and evaluation features best kept distinct.
-
Representation-Guided Discrete Molecular Graph Retrosynthesis
GRG achieves 58.6/77.2/83.4/87.1 top-1/3/5/10 accuracy and 15.5 diversity on USPTO-50k retrosynthesis, outperforming the base generator while reducing training time by 30%.
-
RiT: Vanilla Diffusion Transformers Suffice in Representation Space
A vanilla Diffusion Transformer trained via x-prediction on frozen DINOv2 features reaches FID 1.14 on ImageNet 256x256 with fewer parameters and faster sampling than prior DiT variants.
-
Spatial Gram Alignment for Ultra-High-Resolution Image Synthesis
Spatial Gram Alignment aligns internal self-similarities of LDM features with foundation priors to reconcile global structure and fine details in ultra-high-resolution text-to-image synthesis.
-
Pareto-Enhanced Portrait Generation: Vision-Aligned Text Supervision for Alignment, Realism, and Aesthetics
A feature supervision approach using SigLIP 2 extracts multi-granularity vision-aligned text representations to supervise MM-DiT image branches, pushing the Pareto frontier for portrait generation across alignment, realism, and aesthetics.
-
UniRefiner: Teaching Pre-trained ViTs to Self-Dispose Dross via Contrastive Register
UniRefiner uses contrastive registers and a dual alignment objective to remove three categories of spurious tokens from pre-trained ViTs, yielding up to 9.4% mIoU gains on ADE20K and 22% zero-shot segmentation improvements.
-
Lance: Unified Multimodal Modeling by Multi-Task Synergy
Lance presents a dual-stream mixture-of-experts model with modality-aware positional encoding and staged multi-task training that outperforms prior open-source unified models on image and video generation while keeping strong understanding performance.
-
Resolving Representation Ambiguity in Feedforward Novel View Synthesis Transformer via Semantic-Spatial Decoupling
Decouples semantic and spatial tokens in NVS transformers to resolve representation ambiguity, yielding consistent gains with near-zero added latency.
-
Vision Foundation Models as Generalist Tokenizers for Image Generation
VFMTok builds a generalist image tokenizer on frozen VFMs using adaptive quantization and semantic alignment, delivering gFID 1.36 for autoregressive and 1.25 for continuous generation on ImageNet with 3x faster convergence.
-
GeoFlow: Enforcing Implicit Geometric Consistency in Video Generation
GeoFlow adds a geometry-consistency reward based on rigid camera flow and object appearance preservation, integrated via reinforcement fine-tuning to improve geometric coherence in video generation.
-
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.
-
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.
-
Taming Audio VAEs via Target-KL Regularization
The paper introduces target-KL regularization to train audio VAEs at specific bitrates, enabling rate-distortion curves and comparison to discrete audio codecs for improved text-to-sound generation.
-
Beyond Point-Wise Matching: Structural Representation Alignment for Accelerating Diffusion Transformers
sREPA enforces structural consistency in relational geometry of pre-trained vision features to accelerate DiT training and improve generation quality.
-
Registers Matter for Pixel-Space Diffusion Transformers
Register tokens enhance pixel-space DiT training and output quality via cleaner high-noise feature maps, and a dual-stream design adds further gains with little overhead.
-
Aligning Latent Geometry for Spherical Flow Matching in Image Generation
Projecting VAE latents to a fixed spherical radius and replacing linear interpolation with spherical linear interpolation improves class-conditional ImageNet-256 FID while leaving the diffusion architecture unchanged.
-
One-Step Generative Modeling via Wasserstein Gradient Flows
W-Flow compresses a Wasserstein gradient flow defined via Sinkhorn divergence into a single-step neural generator, reporting 1.29 FID on ImageNet 256x256 with improved mode coverage.
-
PoDAR: Power-Disentangled Audio Representation for Generative Modeling
PoDAR disentangles audio signal power from semantic content in latents using power augmentation and consistency objectives, yielding 2x faster convergence and gains of 0.055 speaker similarity and 0.22 UTMOS when applied to Stable Audio VAE with F5-TTS.
-
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.
-
What Matters for Diffusion-Friendly Latent Manifold? Prior-Aligned Autoencoders for Latent Diffusion
Prior-Aligned AutoEncoders shape latent manifolds with spatial coherence, local continuity, and global semantics to improve latent diffusion, achieving SOTA gFID 1.03 on ImageNet 256x256 with up to 13x faster convergence.
-
SARA: Semantically Adaptive Relational Alignment for Video Diffusion Models
SARA introduces semantic saliency to guide relational alignment in video diffusion models, improving text following and motion quality over prior alignment methods.
-
Toward Better Geometric Representations for Molecule Generative Models
LENSEs improves representation-conditioned molecule generation by jointly training a multi-level representation head, perceptual loss, and REPA alignment on pretrained encoders, yielding 97.28% validity and 98.51% stability on GEOM-DRUG.