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 Baseline reference
X-omni: Reinforcement learning makes discrete autoregressive image generative models great again
Baseline reference. 71% of citing Pith papers use this work as a benchmark or comparison.
29
Pith papers citing it
Baseline
71% of classified citations
hub tools
citation-role summary
baseline 3
background 2
dataset 2
citation-polarity summary
representative citing papers
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.
Geo-Align applies RL with a perceptual reward derived from 3D camera trajectory estimation to improve controllability and fidelity in video generation without paired training data.
MULTITEXTEDIT benchmark reveals that all tested text-in-image editing models show pronounced degradation on non-English languages, especially Hebrew and Arabic, mainly in text accuracy and script fidelity.
TextGround4M supplies 4M prompt-aligned image pairs with layout annotations, enabling autoregressive T2I models to render prompt-specified text more accurately via added span tokens, a new benchmark, and layout-aware metrics.
A self-evolving framework with proposer-solver-generator roles, Solver Token Entropy, and multi-scale internal evaluation improves unified LMMs on understanding and generation tasks using only self-derived consistency signals.
Unison is a new benchmark with unified and decoupled tracks plus Unison-Judge to measure synergy between understanding and generation in multimodal models.
UniAR uses a shared context-visual tokenizer with bitwise quantization and parallel prediction in an autoregressive framework to unify visual understanding and generation, claiming SOTA on generation and editing tasks.
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.
Z-Image Turbo++ narrows the quality gap to 8-step generation via three distillation techniques tailored for the 2-step regime.
GenClaw introduces a three-stage code-driven workflow for agentic image generation that inserts programmatic sketches between linguistic reasoning and pixel synthesis.
TextSculptor supplies an automated data synthesis pipeline yielding 3.2M samples plus a four-task benchmark that raises open-source scene text editing performance.
Generative Visual Grounding creates instance-specific visual proxy images from EEG signals to enhance MLLM understanding of brain activity beyond text-only alignment.
Proposes HT-GRPO with sketch-then-paint staged updates, prompt-conditioned importance ratios, and hierarchical credit assignment for dMLLMs, reporting gains on GenEval and DPG plus quality metrics.
A pixel-space Diffusion Transformer with Unified Transformer architecture unifies image generation, editing, and personalization in an end-to-end model that maps all inputs to a shared token space and scales from 8B to over 200B parameters.
FlashAR accelerates autoregressive image generation up to 22.9x by post-training a pre-trained raster-scan model with a complementary vertical head and dynamic fusion for two-way next-token prediction.
Tuna-2 shows that direct pixel embeddings can replace vision encoders in unified multimodal models, achieving competitive generation and stronger understanding at scale.
LLaDA2.0-Uni unifies multimodal understanding and generation inside one discrete diffusion large language model with a semantic tokenizer, MoE backbone, and diffusion decoder.
Emu3.5 is a native multimodal world model pre-trained on over 10 trillion vision-language tokens with next-token prediction, post-trained via reinforcement learning, and accelerated by Discrete Diffusion Adaptation for efficient interleaved generation and world exploration.
Vega unifies video understanding and generation via shared vocabulary and hybrid autoregressive-diffusion architecture, reporting strong results on VBench and VideoMME.
RSICCLLM introduces a post-training framework with RSICI dataset, difference-aware supervised fine-tuning, and dual-negative preference optimization that claims to outperform much larger models on remote sensing image change captioning.
TMP is a pruning framework that reduces HunyuanImage-3.0 from 80B to 20B parameters (75% reduction) and Z-Image turbo from 6B to 4B with limited quality degradation.
Lens is a 3.8B-parameter text-to-image model that reaches competitive or superior performance to >6B-parameter systems using 19.3% of the training compute of Z-Image through a densely captioned 800M dataset, multi-resolution batching, semantic VAE, strong language encoder, RL fine-tuning, and 4-step
SenseNova-U1 presents native unified multimodal models that match top understanding VLMs while delivering strong performance in image generation, infographics, and interleaved tasks via the NEO-unify architecture.
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.
-
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.
-
Geo-Align: Video Generation Alignment via Metric Geometry Reward
Geo-Align applies RL with a perceptual reward derived from 3D camera trajectory estimation to improve controllability and fidelity in video generation without paired training data.
-
MULTITEXTEDIT: Benchmarking Cross-Lingual Degradation in Text-in-Image Editing
MULTITEXTEDIT benchmark reveals that all tested text-in-image editing models show pronounced degradation on non-English languages, especially Hebrew and Arabic, mainly in text accuracy and script fidelity.
-
TextGround4M: A Prompt-Aligned Dataset for Layout-Aware Text Rendering
TextGround4M supplies 4M prompt-aligned image pairs with layout annotations, enabling autoregressive T2I models to render prompt-specified text more accurately via added span tokens, a new benchmark, and layout-aware metrics.
-
Ask, Solve, Generate: Self-Evolving Unified Multimodal Understanding and Generation via Self-Consistency Rewards
A self-evolving framework with proposer-solver-generator roles, Solver Token Entropy, and multi-scale internal evaluation improves unified LMMs on understanding and generation tasks using only self-derived consistency signals.
-
Unison: Benchmarking Unified Multimodal Models via Synergistic Understanding and Generation
Unison is a new benchmark with unified and decoupled tracks plus Unison-Judge to measure synergy between understanding and generation in multimodal models.
-
Unified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to Unification
UniAR uses a shared context-visual tokenizer with bitwise quantization and parallel prediction in an autoregressive framework to unify visual understanding and generation, claiming SOTA on generation and editing tasks.
-
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.
-
High-Fidelity Two-Step Image Generation via Teacher-Aligned End-to-End Distillation
Z-Image Turbo++ narrows the quality gap to 8-step generation via three distillation techniques tailored for the 2-step regime.
-
GenClaw: Code-Driven Agentic Image Generation
GenClaw introduces a three-stage code-driven workflow for agentic image generation that inserts programmatic sketches between linguistic reasoning and pixel synthesis.
-
TextSculptor: Training and Benchmarking Scene Text Editing
TextSculptor supplies an automated data synthesis pipeline yielding 3.2M samples plus a four-task benchmark that raises open-source scene text editing performance.
-
Visualizing the Invisible: Generative Visual Grounding Empowers Universal EEG Understanding in MLLMs
Generative Visual Grounding creates instance-specific visual proxy images from EEG signals to enhance MLLM understanding of brain activity beyond text-only alignment.
-
Sketch Then Paint: Hierarchical Reinforcement Learning for Diffusion Multi-Modal Large Language Models
Proposes HT-GRPO with sketch-then-paint staged updates, prompt-conditioned importance ratios, and hierarchical credit assignment for dMLLMs, reporting gains on GenEval and DPG plus quality metrics.
-
HiDream-O1-Image: A Natively Unified Image Generative Foundation Model with Pixel-level Unified Transformer
A pixel-space Diffusion Transformer with Unified Transformer architecture unifies image generation, editing, and personalization in an end-to-end model that maps all inputs to a shared token space and scales from 8B to over 200B parameters.
-
FlashAR: Efficient Post-Training Acceleration for Autoregressive Image Generation
FlashAR accelerates autoregressive image generation up to 22.9x by post-training a pre-trained raster-scan model with a complementary vertical head and dynamic fusion for two-way next-token prediction.
-
Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation
Tuna-2 shows that direct pixel embeddings can replace vision encoders in unified multimodal models, achieving competitive generation and stronger understanding at scale.
-
LLaDA2.0-Uni: Unifying Multimodal Understanding and Generation with Diffusion Large Language Model
LLaDA2.0-Uni unifies multimodal understanding and generation inside one discrete diffusion large language model with a semantic tokenizer, MoE backbone, and diffusion decoder.
-
Emu3.5: Native Multimodal Models are World Learners
Emu3.5 is a native multimodal world model pre-trained on over 10 trillion vision-language tokens with next-token prediction, post-trained via reinforcement learning, and accelerated by Discrete Diffusion Adaptation for efficient interleaved generation and world exploration.
-
Bridging Video Understanding and Generation in a Unified Framework
Vega unifies video understanding and generation via shared vocabulary and hybrid autoregressive-diffusion architecture, reporting strong results on VBench and VideoMME.
-
RSICCLLM: A Multimodal Large Language Model for Remote Sensing Image Change Captioning
RSICCLLM introduces a post-training framework with RSICI dataset, difference-aware supervised fine-tuning, and dual-negative preference optimization that claims to outperform much larger models on remote sensing image change captioning.
-
TMP: Tree-structured Mixed-policy Pruning for Large-scale Image Generation and Editing
TMP is a pruning framework that reduces HunyuanImage-3.0 from 80B to 20B parameters (75% reduction) and Z-Image turbo from 6B to 4B with limited quality degradation.
-
Lens: Rethinking Training Efficiency for Foundational Text-to-Image Models
Lens is a 3.8B-parameter text-to-image model that reaches competitive or superior performance to >6B-parameter systems using 19.3% of the training compute of Z-Image through a densely captioned 800M dataset, multi-resolution batching, semantic VAE, strong language encoder, RL fine-tuning, and 4-step
-
SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture
SenseNova-U1 presents native unified multimodal models that match top understanding VLMs while delivering strong performance in image generation, infographics, and interleaved tasks via the NEO-unify architecture.
-
Context Unrolling in Omni Models
Omni is a multimodal model whose native training on diverse data types enables context unrolling, allowing explicit reasoning across modalities to better approximate shared knowledge and improve downstream performance.
-
Qwen-Image Technical Report
Qwen-Image is a foundation model that reaches state-of-the-art results in image generation and editing by combining a large-scale text-focused data pipeline with curriculum learning and dual semantic-reconstructive encoding for editing consistency.
-
ERNIE-Image Technical Report
The paper presents ERNIE-Image, an open-source 8B DiT text-to-image model claiming leading open-source performance and near-commercial results via specialized data construction and DPO alignment.
-
JoyAI-Image: Awaking Spatial Intelligence in Unified Multimodal Understanding and Generation
JoyAI-Image unifies visual understanding and generation via an MLLM-MMDiT architecture with spatial training signals to reach competitive benchmark performance and stronger spatial intelligence.
- Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer