The Year in AI Papers
The year/Topics/Image generation and editing

Research collection

Image generation and editing

Text-to-image and image-editing models: diffusion transformers, autoregressive image models, and image generation foundation models.

Papers
53
Research labs
4
Official code
41

1–50 of 53 papers in this collection

01

Independent research

Rethinking Classifier-Free Guidance in On-Policy Diffusion Distillation

This paper analyzes on-policy distillation (OPD) for diffusion models under classifier-free guidance (CFG). The authors show that the naive objective of matching CFG-composed velocities is under-identified at the branch level, allowing positive- and negative-branch errors to cancel. They identify a failure mode, Negative Branch Asymmetry (NBA), which…

Bingnan Li, Haozhe Wang, Haozhong Xiong, Fangtai Wu, et al.
Published
Jul 2026
Citations
0
Code
8 stars
Read summary Original
02

Independent research

Mage-Flow: An Efficient Native-Resolution Foundation Model for Image Generation and Editing

Mage-Flow is a compact 4B-parameter generative stack from Microsoft for efficient text-to-image generation and instruction-based image editing. It comprises Mage-VAE, a lightweight latent tokenizer using one-step diffusion-style encoding/decoding with anchor-latent regularization, and a Native-Resolution Multimodal Diffusion Transformer (NR-MMDiT) trained…

Xinjie Zhang, Peng Zhang, Shicheng Zheng, Jinghao Guo, et al.
Published
Jul 2026
Citations
1
Code
Not linked
Read summary Original
03

Independent research

Boogu-Image-0.1: Boosting Open Agentic Multimodal Generation via Understanding under a Minimal Budget

Boogu-Image-0.1 is an open-source family of unified multimodal understanding and generation models (Base, Turbo, Edit, Edit-Turbo) that achieves competitive performance in text-to-image generation, fast inference, instruction-based editing, and bilingual text rendering. The authors argue that strengthening understanding—via a stronger multimodal encoder…

Guoxuan Chen, Chufeng Xiao, Haoran Yang, Siyue Xie, et al.
Published
Jul 2026
Citations
0
Code
Not linked
Read summary Original
04

Independent research

DanceOPD: On-Policy Generative Field Distillation

DanceOPD is an on-policy generative field distillation framework for flow-matching image generation models, designed to compose multiple capabilities (e.g., text-to-image, local editing, global editing) into a single student model. The method treats each frozen capability source as a velocity field over a shared state space and addresses three key…

Wei Zhou, Xiongwei Zhu, Zelin Xu, Bo Dong, et al.
Published
Jun 2026
Citations
2
Code
395 stars
Read summary Original
05

Independent research

Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance

Moebius is a highly efficient lightweight image inpainting framework that rivals the generation quality of 10B-level industrial models like FLUX.1-Fill-Dev while using only 0.22B parameters (less than 2% of FLUX's 11.9B) and delivering over 15× faster total inference time. To overcome the representation bottleneck from extreme structural compression, the…

Kangsheng Duan, Ziyang Xu, Wenyu Liu, Xiaohu Ruan, et al.
Published
Jun 2026
Citations
0
Code
508 stars
Read summary Original
06

Independent research

InterleaveThinker: Reinforcing Agentic Interleaved Generation

InterleaveThinker is a multi-agent framework that endows existing image generators with interleaved text-image generation capabilities, addressing visual over-reliance and step-wise error accumulation in Unified Multimodal Models (UMMs). It uses a Planner agent to pre-plan the full instruction sequence, a Generator (e.g., FLUX.2-klein-9B) to execute steps,…

Dian Zheng, Harry Lee, Manyuan Zhang, Kaituo Feng, et al.
Published
Jun 2026
Citations
0
Code
208 stars
Read summary Original
07

arXiv.org

Crafter: A Multi-Agent Harness for Editable Scientific Figure Generation from Diverse Inputs

The paper introduces CRAFTER, a multi-agent harness for generating scientific figures from diverse inputs, and CRAFTEDITOR, which converts raster outputs into editable SVGs. CRAFTER uses five agents (intent reasoner, plan generator, critic, specification refiner, convergence judge) sharing an evolving specification, with mechanisms for diversity-driven…

Haozhe Zhao, Shuzheng Si, Zhenhailong Wang, Zheng Wang, et al.
Published
May 2026
Citations
1
Code
149 stars
Read summary Original
08

arXiv.org

CollectionLoRA: Collecting 50 Effects in 1 LoRA via Multi-Teacher On-Policy Distillation

CollectionLoRA is a multi-teacher on-policy distillation framework that consolidates up to 50 diverse visual effects and few-step generation capabilities into a single LoRA, addressing storage overhead, routing latency, and parameter conflicts in conventional multi-LoRA pipelines. The method introduces three key components: Probabilistic Dual-Stream…

Fangtai Wu, Hailong Guo, Shijie Huang, Jiayi Song, et al.
Published
May 2026
Citations
2
Code
29 stars
Read summary Original
09

arXiv.org

Lens: Rethinking Training Efficiency for Foundational Text-to-Image Models

Lens is a 3.8B-parameter text-to-image model that achieves performance competitive with larger state-of-the-art models while using significantly less training compute, requiring only about 19.3% of the compute used by Z-Image. Its efficiency stems from maximizing data information density per batch via the Lens-800M dataset of densely captioned image-text…

Dong Chen, Fangyun Wei, Ziyu Wan, Dongdong Chen, et al.
Published
May 2026
Citations
1
Code
Not linked
Read summary Original
10

arXiv.org

Rethinking Cross-Layer Information Routing in Diffusion Transformers

This paper investigates cross-layer information routing in Diffusion Transformers (DiTs), identifying three symptoms of standard residual connections: forward magnitude inflation, backward gradient decay, and block-wise redundancy. The authors propose Diffusion-Adaptive Routing (DAR), a drop-in replacement that uses learnable, timestep-adaptive,…

Chao Xu, Maohua Li, Qirui Li, Yixuan Xu, et al.
Published
May 2026
Citations
2
Code
Not linked
Read summary Original
11

Qwen

Qwen-Image-VAE-2.0 Technical Report

Qwen-Image-VAE-2.0 is a suite of high-compression image VAEs (f16 and f32) designed to overcome the trade-off between compression ratio, reconstruction fidelity, and diffusability. The architecture uses Global Skip Connections (GSC) to preserve fine details, expanded latent channels, and an attention-free, asymmetric encoder-decoder backbone for…

Zekai Zhang, Deqing Li, Kuan Cao, Yujia Wu, et al.
Published
May 2026
Citations
1
Code
69 stars
Read summary Original
12

Qwen

Qwen-Image-2.0 Technical Report

Qwen-Image-2.0 is an image generation foundation model that unifies text-to-image (T2I) generation and instruction-based image editing in a single framework. It addresses challenges in ultra-long text rendering (up to 1K tokens), multilingual typography, high-resolution photorealism (native 2K), complex instruction following, and inference efficiency. The…

Bing Zhao, Chenfei Wu, Deqing Li, Hao Meng, et al.
Published
May 2026
Citations
8
Code
Not linked
Read summary Original
13

arXiv.org

Tstars-Tryon 1.0: Robust and Realistic Virtual Try-On for Diverse Fashion Items

Tstars-Tryon 1.0, developed by the Pailitao Team at Alibaba Group, is a commercial-scale virtual try-on system designed for robustness, realism, versatility, and efficiency. It handles challenging real-world cases like extreme poses, lighting variations, and motion blur, while preserving garment details and avoiding synthetic artifacts. The system supports…

Mengting Chen, Zhengrui Chen, Yongchao Du, Zuan Gao, et al.
Published
Apr 2026
Citations
2
Code
Not linked
Read summary Original
14

arXiv.org

Extending One-Step Image Generation from Class Labels to Text via Discriminative Text Representation

This paper extends MeanFlow, a one-step generation framework originally designed for class-label conditioning, to flexible text-to-image (T2I) generation. The authors find that directly integrating LLM-based text encoders into MeanFlow with standard training fails, and they identify that high-quality text representations must possess strong semantic…

Chenxi Zhao, Chen Zhu, Xiaokun Feng, Aiming Hao, et al.
Published
Apr 2026
Citations
2
Code
109 stars
Read summary Original
15

arXiv.org

Elucidating the SNR-t Bias of Diffusion Probabilistic Models

This paper identifies a Signal-to-Noise Ratio-timestep (SNR-t) bias in Diffusion Probabilistic Models (DPMs), where during inference the SNR of denoised samples becomes misaligned with their timestep due to accumulated prediction and discretization errors. The authors provide empirical evidence and theoretical proof showing that reverse-process samples…

Meng Yu, Lei Sun, Jianhao Zeng, Xiangxiang Chu, et al.
Published
Apr 2026
Citations
2
Code
120 stars
Read summary Original
16

arXiv.org

RationalRewards: Reasoning Rewards Scale Visual Generation Both Training and Test Time

The paper introduces RationalRewards, a reasoning-based reward model for visual generation that produces structured, multi-dimensional critiques before assigning scores, unlike traditional scalar reward models. It is trained using Preference-Anchored Rationalization (PARROT), a variational framework that recovers rationales from preference data via…

Haozhe Wang, Cong Wei, Weiming Ren, Jiaming Liu, et al.
Published
Apr 2026
Citations
7
Code
56 stars
Read summary Original
17

Meta AI

Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoning

This paper introduces process-driven image generation, a multi-step paradigm that decomposes text-to-image synthesis into an interleaved reasoning trajectory of textual planning and visual generation. The method uses a recurring four-stage cycle: Plan, Sketch, Inspect, and Refine, where the model generates incremental instructions and scene descriptions,…

Lei Zhang, Junjiao Tian, Zhipeng Fan, Kunpeng Li, et al.
Published
Apr 2026
Citations
3
Code
Not linked
Read summary Original
18

arXiv.org

GEMS: Agent-Native Multimodal Generation with Memory and Skills

GEMS is a framework for multimodal generation that uses an agent-native approach to improve performance on complex instructions and specialized tasks. It has three main parts: an Agent Loop that iteratively refines generation through planning, decomposition, generation, verification, and refinement; Agent Memory that stores a persistent, hierarchically…

Zefeng He, Siyuan Huang, Xiaoye Qu, Yafu Li, et al.
Published
Mar 2026
Citations
8
Code
142 stars
Read summary Original
19

arXiv.org

PixelSmile: Toward Fine-Grained Facial Expression Editing

PixelSmile is a diffusion-based framework for fine-grained facial expression editing, addressing semantic overlap between expressions like fear-surprise and anger-disgust. The authors construct the Flex Facial Expression (FFE) dataset with 60,000 images (real and anime) annotated with continuous 12-dimensional affective scores, and establish FFE-Bench to…

Jiabin Hua, Hengyuan Xu, Aojie Li, Wei Cheng, et al.
Published
Mar 2026
Citations
0
Code
Not linked
Read summary Original
20

arXiv.org

RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Models

RealRestorer is an open-source image restoration model designed to handle diverse real-world degradations, including blur, rain, noise, low-light, moiré patterns, haze, compression artifacts, reflection, and flare. The authors construct a large-scale dataset with a synthesis pipeline that combines synthetic and real-world degradation data, and they…

Yufeng Yang, Xianfang Zeng, Zhangqi Jiang, Fukun Yin, et al.
Published
Mar 2026
Citations
3
Code
Not linked
Read summary Original
21

arXiv.org

Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration

The paper introduces Calibri, a parameter-efficient method to enhance Diffusion Transformers (DiTs) by calibrating block outputs with learned scaling parameters. The authors show that selectively disabling or re-weighting DiT blocks can improve generation quality, leading to a black-box optimization problem solved via CMA-ES, modifying only ~10^2…

Danil Tokhchukov, Aysel Mirzoeva, Andrey Kuznetsov, Konstantin Sobolev
Published
Mar 2026
Citations
0
Code
59 stars
Read summary Original
22

arXiv.org

Enhancing Spatial Understanding in Image Generation via Reward Modeling

The paper introduces a method to improve spatial understanding in text-to-image generation using reward modeling. The authors construct the SpatialReward-Dataset, containing over 80,000 adversarial preference pairs, where each pair consists of an image correctly depicting complex spatial relationships and a perturbed image violating some relationships,…

Zhenyu Tang, Chaoran Feng, Yufan Deng, Jie Wu, et al.
Published
Feb 2026
Citations
0
Code
86 stars
Read summary Original
23

Google DeepMind

Unified Latents (UL): How to train your latents

Unified Latents (UL) is a framework for learning latent representations regularized by a diffusion prior and decoded by a diffusion model. The encoder outputs a deterministic latent, which is noised to a fixed minimum noise level (log-SNR of 5), linking the encoder's output noise to the prior's precision. The training objective combines a diffusion prior…

Jonathan Heek, Emiel Hoogeboom, Thomas Mensink, Tim Salimans
Published
Feb 2026
Citations
15
Code
Not linked
Read summary Original
24

arXiv.org

BitDance: Scaling Autoregressive Generative Models with Binary Tokens

BitDance is a scalable autoregressive image generation model that predicts binary visual tokens instead of codebook indices, scaling the vocabulary to 2^256 states. It introduces a binary diffusion head to sample from this large discrete space, and a next-patch diffusion method for parallel multi-token prediction. On ImageNet 256×256, BitDance achieves an…

Yuang Ai, Jiaming Han, Shaobin Zhuang, Weijia Mao, et al.
Published
Feb 2026
Citations
5
Code
481 stars
Read summary Original
25

arXiv.org

DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing

DeepGen 1.0 is a lightweight 5B-parameter unified multimodal model (3B VLM + 2B DiT) for image generation and editing, achieving performance competitive with or surpassing much larger models. It introduces Stacked Channel Bridging (SCB), which fuses features from six uniformly distributed VLM layers with learnable 'think tokens' to provide the DiT with…

Dianyi Wang, Ruihang Li, Feng Han, Chaofan Ma, et al.
Published
Feb 2026
Citations
15
Code
584 stars
Read summary Original
26

arXiv.org

UniReason 1.0: A Unified Reasoning Framework for World Knowledge Aligned Image Generation and Editing

UniReason is a unified framework for text-to-image (T2I) generation and image editing, built on the Bagel architecture. It addresses limitations of existing methods by incorporating world knowledge-enhanced textual reasoning before synthesis and fine-grained editing-like visual refinement after initial generation. The framework uses two complementary…

Dianyi Wang, Chaofan Ma, Feng Han, Size Wu, et al.
Published
Feb 2026
Citations
6
Code
144 stars
Read summary Original
27

arXiv.org

Everything in Its Place: Benchmarking Spatial Intelligence of Text-to-Image Models

The paper introduces SpatialGenEval, a benchmark for evaluating the spatial intelligence of text-to-image (T2I) models. It uses 1,230 long, information-dense prompts across 25 real-world scenes, each integrating 10 spatial sub-domains (object, attribute, position, orientation, layout, comparison, proximity, occlusion, motion, causal) and paired with 10…

Zengbin Wang, Xuecai Hu, Yong Wang, Feng Xiong, et al.
Published
Jan 2026
Citations
8
Code
132 stars
Read summary Original
28

arXiv.org

VIBE: Visual Instruction Based Editor

VIBE is a compact, high-throughput instruction-based image editing pipeline that combines a 2B-parameter Qwen3-VL model for instruction interpretation with a 1.6B-parameter Sana1.5 diffusion model for image generation. The architecture uses channel-wise concatenation for reference image guidance and learnable meta-tokens processed by a lightweight…

Grigorii Alekseenko, Aleksandr Gordeev, Irina Tolstykh, Bulat Suleimanov, et al.
Published
Jan 2026
Citations
2
Code
55 stars
Read summary Original
29

arXiv.org

The Prism Hypothesis: Harmonizing Semantic and Pixel Representations via Unified Autoencoding

The paper introduces the Prism Hypothesis, which posits that multimodal data can be understood through a shared frequency spectrum: semantic encoders capture low-frequency components (abstract meaning), while pixel encoders retain high-frequency details (fine texture). This is supported by experiments showing that text-image retrieval relies on low…

Weichen Fan, Haiwen Diao, Quan Wang, Dahua Lin, et al.
Published
Dec 2025
Citations
15
Code
209 stars
Read summary Original
30

Qwen

Qwen-Image-Layered: Towards Inherent Editability via Layer Decomposition

Qwen-Image-Layered is an end-to-end diffusion model that decomposes a single RGB image into multiple semantically disentangled RGBA layers, enabling consistent image editing where each layer can be independently manipulated. The model introduces three key components: an RGBA-VAE that unifies latent representations for RGB and RGBA images, a VLD-MMDiT…

Shengming Yin, Zekai Zhang, Zecheng Tang, Kaiyuan Gao, et al.
Published
Dec 2025
Citations
24
Code
2.1K stars
Read summary Original
31

MiniMax

Towards Scalable Pre-training of Visual Tokenizers for Generation

The paper introduces VTP, a visual tokenizer pre-training framework that integrates image-text contrastive learning, self-supervised learning (MIM and self-distillation), and reconstruction losses to address the 'pre-training scaling problem' in latent diffusion models. The authors argue that reconstruction-only training biases the latent space toward…

Jingfeng Yao, Yuda Song, Yucong Zhou, Xinggang Wang
Published
Dec 2025
Citations
23
Code
497 stars
Read summary Original
32

arXiv.org

TwinFlow: Realizing One-step Generation on Large Models with Self-adversarial Flows

TWINFLOW is a framework for training one-step generative models on large multimodal models, addressing the inefficiency of multi-step diffusion and flow matching models that require 40-100 NFEs. Existing few-step methods either rely on auxiliary trained models (e.g., GAN discriminators) or frozen teachers, causing instability and memory overhead, or…

Zhenglin Cheng, Peng Sun, Jianguo Li, Tao Lin
Published
Dec 2025
Citations
16
Code
537 stars
Read summary Original
33

arXiv.org

Envision: Benchmarking Unified Understanding & Generation for Causal World Process Insights

The paper introduces Envision, a benchmark for evaluating text-to-image (T2I) and unified multimodal models (UMMs) on causal, multi-image event generation. It addresses the limitation of static single-image benchmarks by proposing chained text-to-multi-image generation with 1,000 four-stage prompts across six scientific and humanities domains.…

Juanxi Tian, Siyuan Li, Conghui He, Lijun Wu, et al.
Published
Dec 2025
Citations
1
Code
32 stars
Read summary Original
34

arXiv.org

Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer

Z-Image is a 6B-parameter image generation foundation model from Alibaba Group, built on a Scalable Single-Stream Diffusion Transformer (S3-DiT). It challenges the 'scale-at-all-costs' paradigm by optimizing data infrastructure, architecture, training, and inference. The full training workflow costs 314K H800 GPU hours (~$628K). Z-Image-Turbo, a distilled…

Z-Image Team, Huanqia Cai, Sihan Cao, Ruoyi Du, et al.
Published
Nov 2025
Citations
206
Code
12K stars
Read summary Original
35

arXiv.org

DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generation

DeCo is a frequency-decoupled pixel diffusion framework for end-to-end image generation. It addresses the challenge of pixel diffusion models jointly modeling high-frequency signals and low-frequency semantics in a single diffusion transformer (DiT), which slows training and inference. DeCo uses a DiT to model low-frequency semantics from downsampled…

Zehong Ma, Longhui Wei, Shuai Wang, Shiliang Zhang, et al.
Published
Nov 2025
Citations
37
Code
238 stars
Read summary Original
36

arXiv.org

Kandinsky 5.0: A Family of Foundation Models for Image and Video Generation

Kandinsky 5.0 is a family of foundation models for high-resolution image and video synthesis, comprising three model line-ups: Image Lite (6B parameters), Video Lite (2B), and Video Pro (19B). The models are built on a unified latent diffusion architecture with a CrossDiT backbone and trained using flow matching. Key innovations include the NABLA sparse…

Vladimir Arkhipkin, Vladimir Korviakov, Nikolai Gerasimenko, Denis Parkhomenko, et al.
Published
Nov 2025
Citations
9
Code
805 stars
Read summary Original
37

arXiv (Cornell University)

Back to Basics: Let Denoising Generative Models Denoise

The paper argues that denoising diffusion models should directly predict clean images (x-prediction) rather than noise (epsilon-prediction) or velocity (v-prediction), as natural data lies on a low-dimensional manifold while noised quantities do not. The authors propose 'Just image Transformers' (JiT), a plain Vision Transformer applied to large pixel…

Tianhong Li, Kaiming He
Published
Nov 2025
Citations
4
Code
2.5K stars
Read summary Original
38

arXiv.org

One Small Step in Latent, One Giant Leap for Pixels: Fast Latent Upscale Adapter for Your Diffusion Models

The paper introduces the Latent Upscaler Adapter (LUA), a lightweight module that performs super-resolution directly on a diffusion model's latent code before VAE decoding, enabling high-resolution image synthesis without retraining the generator or adding diffusion stages. LUA uses a shared SwinIR-style backbone with scale-specific pixel-shuffle heads for…

Aleksandr Razin, Danil Kazantsev, Ilya Makarov
Published
Nov 2025
Citations
1
Code
34 stars
Read summary Original
39

arXiv.org

A Style is Worth One Code: Unlocking Code-to-Style Image Generation with Discrete Style Space

The paper introduces CoTyle, the first open-source framework for code-to-style image generation, where a numerical style code controls the visual style of generated images. Existing methods rely on textual prompts, reference images, or LoRA fine-tuning, which suffer from poor consistency, limited creativity, or complex representations. CoTyle trains a…

Huijie Liu, Shuhao Cui, Haoxiang Cao, Shuai Ma, et al.
Published
Nov 2025
Citations
0
Code
222 stars
Read summary Original
40

arXiv.org

FARMER: Flow AutoRegressive Transformer over Pixels

FARMER is a novel end-to-end generative framework that unifies Normalizing Flows (NF) and Autoregressive (AR) models for tractable likelihood estimation and high-quality image synthesis directly from raw pixels. It employs an invertible autoregressive flow to transform images into latent sequences, whose distribution is modeled implicitly by an AR model.…

Guangting Zheng, Qinyu Zhao, Tao Yang, Fei Xiao, et al.
Published
Oct 2025
Citations
13
Code
Not linked
Read summary Original
41

arXiv.org

UniGenBench++: A Unified Semantic Evaluation Benchmark for Text-to-Image Generation

UniGenBench++ is a unified semantic evaluation benchmark for text-to-image (T2I) generation, addressing limitations in existing benchmarks such as coarse evaluation dimensions, lack of prompt diversity, and missing multilingual support. It comprises 600 prompts hierarchically organized into 5 primary themes and 20 subthemes, covering 10 primary and 27…

Yibin Wang, Zhimin Li, Yuhang Zang, Jiazi Bu, et al.
Published
Oct 2025
Citations
20
Code
139 stars
Read summary Original
42

arXiv.org

PICABench: How Far Are We from Physically Realistic Image Editing?

PICABench is a benchmark for evaluating physical realism in image editing, addressing the gap where existing models and benchmarks focus on semantic fidelity but overlook physical effects like shadows, reflections, and state transitions. It categorizes physical consistency into three dimensions—Optics, Mechanics, and State Transition—spanning eight…

Yuandong Pu, Le Zhuo, Songhao Han, Jinbo Xing, et al.
Published
Oct 2025
Citations
10
Code
39 stars
Read summary Original
43

arXiv.org

WithAnyone: Towards Controllable and ID Consistent Image Generation

The paper introduces WithAnyone, a diffusion-based model for controllable and identity-consistent image generation, addressing the 'copy-paste' artifact where models replicate reference faces instead of preserving identity across variations. The authors construct MultiID-2M, a large-scale dataset with 500k group photos and paired references for 3k…

Hengyuan Xu, Wei Cheng, Peng Xing, Yixiao Fang, et al.
Published
Oct 2025
Citations
14
Code
573 stars
Read summary Original
44

Independent research

There is No VAE: End-to-End Pixel-Space Generative Modeling via Self-Supervised Pre-training

The paper introduces EPG, a two-stage training framework for pixel-space diffusion and consistency models that closes the performance gap with latent-space methods. Stage one pre-trains an encoder using a contrastive loss and a representation consistency loss, aligning features of noisy images along deterministic sampling trajectories. Stage two fine-tunes…

Jiachen Lei, Keli Liu, Julius Berner, Haiming Yu, et al.
Published
Oct 2025
Citations
15
Code
153 stars
Read summary Original
45

arXiv.org

Diffusion Transformers with Representation Autoencoders

The paper introduces Representation Autoencoders (RAEs), which replace the traditional VAE in latent diffusion models with a frozen pretrained representation encoder (e.g., DINOv2, SigLIP2, MAE) paired with a trained decoder. RAEs provide both high-quality reconstructions and semantically rich latent spaces, outperforming SD-VAE in reconstruction FID…

Boyang Zheng, Nanye Ma, Shengbang Tong, Saining Xie
Published
Oct 2025
Citations
221
Code
2K stars
Read summary Original
46

arXiv.org

DreamOmni2: Multimodal Instruction-based Editing and Generation

DreamOmni2 introduces two new tasks: multimodal instruction-based editing and generation, which accept both text and image instructions and support concrete objects as well as abstract attributes (e.g., texture, pose, style). The authors propose a three-stage data synthesis pipeline: (1) a feature mixing scheme in attention to generate paired images…

Bin Xia, Bohao Peng, Yuechen Zhang, Junjia Huang, et al.
Published
Oct 2025
Citations
34
Code
2K stars
Read summary Original
47

arXiv.org

OpenGPT-4o-Image: A Comprehensive Dataset for Advanced Image Generation and Editing

OpenGPT-4o-Image is a large-scale dataset for training unified multimodal models in image generation and editing. It contains 80,000 instruction-image pairs across 11 major domains and 51 subtasks, built using a hierarchical taxonomy and an automated pipeline that leverages GPT-4o. The taxonomy covers five generation modules (Style Control, Complex…

Zhihong Chen, Xuehai Bai, Yang Shi, Chaoyou Fu, et al.
Published
Sep 2025
Citations
24
Code
33 stars
Read summary Original
48

arXiv.org

Seedream 4.0: Toward Next-generation Multimodal Image Generation

Seedream 4.0 is a multimodal image generation system by ByteDance Seed that unifies text-to-image (T2I) synthesis, image editing, and multi-image composition in a single framework. It uses an efficient diffusion transformer (DiT) with a high-compression VAE, reducing image tokens and enabling native 1K-4K resolution generation. The model is pretrained on…

Team Seedream, :, Yunpeng Chen, Yu Gao, et al.
Published
Sep 2025
Citations
226
Code
Not linked
Read summary Original
49

arXiv.org

USO: Unified Style and Subject-Driven Generation via Disentangled and Reward Learning

The paper introduces USO, a unified framework for style-driven and subject-driven image generation, arguing these tasks are complementary and can be jointly optimized through cross-task co-disentanglement. The authors construct a large-scale triplet dataset (content, style, stylized images) using a subject-for-style curation framework with stylization and…

Shaojin Wu, Mengqi Huang, Yufeng Cheng, Wenxu Wu, et al.
Published
Aug 2025
Citations
29
Code
1.2K stars
Read summary Original
50

arXiv.org

NextStep-1: Toward Autoregressive Image Generation with Continuous Tokens at Scale

NextStep-1 is a 14B autoregressive model for text-to-image generation that uses continuous image tokens instead of discrete VQ tokens, paired with a 157M flow matching head. It is initialized from Qwen2.5-14B and trained with next-token prediction, combining cross-entropy loss for text and flow matching loss for images. The image tokenizer, fine-tuned from…

NextStep Team, Chunrui Han, Guopeng Li, Jingwei Wu, et al.
Published
Aug 2025
Citations
52
Code
694 stars
Read summary Original