Latest Text-to-Image Research Papers

The newest Text-to-Image papers from across the field — arXiv, NeurIPS, CVPR, Nature, and more — refreshed daily and ranked by relevance. Distill AI tracks Text-to-Image so you don’t have to: get the standout work delivered to your inbox every morning, with 2-sentence summaries and the option to chat with any paper.

Recent papers

• VPR-Space: A Benchmark for Visual Place Recognition in Space Analog Domains
  Richard Fechner, Peer Neubert · arXiv · Oct 26, 2026

  Visual Place Recognition (VPR) is an important means for loop closure detection in visual Simultaneous Localization and Mapping (SLAM). While state-of-the-art VPR and SLAM approaches excel in urban and feature-rich natural environments, the…

• SpectralCanary: Embarrassingly Simple Low-Frequency Traces for T2I Dataset-Use Auditing
  Sy-Tuyen Ho, Anirudh Satheesh, Soham Chawathe, Huy Nghiem et al. · arXiv · Oct 22, 2026

  Auditing whether a text-to-image (T2I) model was fine-tuned on a particular image collection is difficult: a model can absorb collection-level visual characteristics without reproducing training images. Proactive dataset-use auditing plants…

• Compositionality with Primitives: A Unified View from Euclidean to Hyperbolic Vision-Language Models
  Puru Vaish, Meelad Dashti, Christoph Brune, Pascal Mettes et al. · arXiv · Oct 21, 2026

  Compositional zero-shot learning (CZSL) aims to recognize novel combinations of previously observed objects and attributes. Prior work showed that large-scale vision-language models encode decomposable semantic structures that can be recomp…

• Continual Concept Personalization in Diffusion Models via Residual-Space Adaptation
  Yuzhi Liu, Enis Simsar, Giulia Lanzillotta, Thomas Hofmann · arXiv · Oct 20, 2026

  Text-to-image personalization aims to equip pretrained diffusion models with the ability to learn user-specific concepts from only a few reference images. Existing methods, however, typically train each concept independently or assume that …

• Copyright Leakage Detection in AI Generated Videos
  Shaif Chowdhury, Moshiur Rahman Tonmoy · arXiv · Oct 20, 2026

  Video generation models can create realistic-looking video footage from images and text prompts. Some of the recent advances have raised concerns about the reproduction of copyrighted visual content. However, copyright leakage detection in …

• Text-to-Image Generative Modeling Without Language Pretraining
  Keshigeyan Chandrasegaran, Yue Zhao, Armin W. Thomas, Ehsan Adeli et al. · arXiv · Oct 19, 2026

  Modern text-to-image (T2I) systems use internet-scale language pretraining for text-conditioned image generation. We ask whether language pretraining is necessary for strong T2I generation. We introduce Halcyon, a family of early-fusion mul…

• GenOlympiad: Benchmarking Frontier Image Models at Their Capability Boundaries
  Xi Hu, Martin Bell, Liang Zheng · arXiv · Oct 19, 2026

  Text-to-image models have made tremendous progress and can produce high-quality images. However, a question which would be asked in any other discipline is where the frontier models fall short. We introduce GenOlympiad, a benchmark which pr…

• PreciseEdit: Strong Yet Stable Semantic Transport for One-Step Image Editing
  Jingyi Wang, Xiao Lu · arXiv · Oct 16, 2026

  Recent one-step text-to-image (T2I) models enable efficient image synthesis, opening up new possibilities for interactive Web-based image editing. However, existing one-step editing methods often construct a global transport field through s…

• Not Just a Subject: Capturing Adaptive Scene-Specific Variance for Visual Storytelling
  SeungJu Cha, Ye-Chan Kim, Kwanyoung Lee, Dong-Jin Kim · arXiv · Oct 15, 2026

  Recent advancements in diffusion models have significantly improved text-to-image generation, enabling their application to visual storytelling, where coherent image sequences are generated from narrative prompts. However, existing methods …

• Beyond Pairwise Repulsion: Collective Diversity via Determinant Maximization in Diffusion Sampling
  Kwanyoung Lee, SeungJu Cha, Yebin Ahn, Sungho Koh et al. · arXiv · Oct 15, 2026

  Strong guidance in text-to-image (T2I) diffusion models concentrates sampling toward high-density modes of the conditional distribution, collapsing independently sampled noise vectors onto nearly identical visual prototypes. This limits uti…

• SCOPE: Semantic Bounded Credit Optimization for Prompt Engineering in Text-to-image Generation
  Yufei Jiao, Jinfa Huang, Zeyue Xue, Tianyu Xie et al. · arXiv · Oct 15, 2026

  Reinforcement learning (RL) enables language models to improve text-to-image (T2I) generation by rewriting prompts for a frozen image generator. Existing methods typically apply a single signal derived from the image to individual subword t…

• Dog-Breed Text-to-image Flow Matching from Pretraining to Fine-Tuning on 4 Consumer GPUs
  Reyhaneh Esmailizadeh, Xingjian Leng, Zhanhao Liang, Liang Zheng · arXiv · Oct 13, 2026

  We generally perceive that training text-to-image diffusion models requires heavy compute: it is true if we aim at a comprehensive frontier model. This understanding prevents people from training their own models when getting started in the…

• Harmful Content Generation in Text-to-Image Models: Capabilities and Moderation Limitations
  Paschalis Giakoumoglou, Manos Schinas, Symeon Papadopoulos · arXiv · Oct 5, 2026

  Text-to-image generative models can produce highly realistic imagery but also raise concerns about harmful misuse. While safety mechanisms exist, systematic evaluations of their effectiveness against realistic attacks remain limited. We pre…

• DEPICT: Scoring Text-to-Image Alignment by Answer Agreement
  Vasco Ramos, Sandra Godinho Silva, Joao Magalhaes, Ricardo Rei et al. · arXiv · Oct 2, 2026

  Image-text alignment is a core problem in computer vision with applications in caption evaluation, hallucination detection, data curation, and the benchmarking of text-to-image (T2I) generators. As T2I models improve, benchmarking has becom…

• ProgressNet: Sketching and Prompting with a Frozen Text-to-Image Model
  Arkaprabha Basu, Chaitat Utintu, Yi-Zhe Song · arXiv · Oct 2, 2026

  Humans draw progressively: a few strokes, a look at the result, a stroke erased, a prompt revised. Image generators do not work this way. They typically take a finished sketch and produce the image in a single pass, so every edit starts the…

• Form and Void: Entangled Composition through an Autonomous AI Agent
  Shiwen Wang, Jian Yang, Xu Wang, Xincan Wang et al. · CVPR · Oct 1, 2026

  Positive and negative space is a fundamental principle in visual composition, supporting visually coherent forms and layered semantic relationships. Generating such compositions is challenging because it requires coordinated control over tw…

• Continual Concept Erasure in Diffusion Models by Suppressing Cross-Edit Interference
  Yongliang Wu, Haori Lu, Jinqi Luo, Wei Cao et al. · arXiv · Oct 1, 2026

  Concept erasure removes copyright-protected, privacy-sensitive, or otherwise undesirable concepts from pretrained text-to-image diffusion models to support content governance and compliance. As erasure requests arrive over time, models must…

• Looped Diffusion Transformer
  Yong Xien Chng, Tianyi Chen, Wenwen Tong, Haiwen Diao et al. · arXiv · Sep 30, 2026

  Improving text-to-image models has traditionally relied on increasing model size or the number of denoising steps. In this work, we explore an alternative way to scale computation by repeatedly running shared Transformer blocks within each …

• DMA$^2$: Pixel-space Distribution Matching with Adversarial and Anchor Losses
  Xin Lin, Zhifei Zhang, Yuqian Zhou, Haitian Zheng et al. · arXiv · Sep 29, 2026

  Distribution matching distillation (DMD) provides a general framework for few-step diffusion generation, but its modern text-to-image instantiations have been developed primarily around latent diffusion. It therefore overlooks key propertie…

• Learning Native Reflection in Unified Models with Interleaved Reinforcement Learning
  Yijia Fan, Ziqi Huang, Zhongang Cai, Yan Li et al. · arXiv · Sep 28, 2026

  Unified multimodal models can both look at and render images, so in principle they can repair their own generations: diagnose what an image gets wrong, revise it, observe the result, and diagnose again. Whether a revision helps is known onl…

• FreeFlow: A Bias-free Hierarchical Transformer for Optical Flow Estimation
  Vladislav Bargatin, Alexander Yakovenko, Khaled Abud, Dmitriy Vatolin · arXiv · Sep 10, 2026

  Optical flow methods typically rely on task-specific inductive biases, such as correlation volumes, feature warping, and iterative refinement, among others, to reach high accuracy. While effective, such biases constrain the model to predefi…

• GDB-Reward: From Evaluation Metrics to Training Rewards for Graphic Design
  Adrienne Deganutti, Purvanshi Mehta, Simon Hadfield, Andrew Gilbert · arXiv · Sep 2, 2026

  Text-to-image models excel at natural image synthesis but struggle with graphic design, where success depends on satisfying precise constraints on typography, layout, color, and visual communication. While prompt optimization offers an attr…

• Genesis: A Generative Engine for Hierarchical Satellite Image Synthesis
  Subash Khanal, Yangzhi Cui, Daniel Cher, Eric Xing et al. · arXiv · Sep 2, 2026

  Earth observation is fundamentally multi-scale; geospatial tasks span varied resolutions, and satellite imagery is organized into cascading tile pyramids that nest fine detail within wide coverage. Current generative models of satellite ima…

• SpatialGuard: Harness-Guided Verifiable Spatial Reasoning for Text-to-Image Generation
  Ziyun Qian, Zizhi Chen, Yizhou Liu, Mingyang Sun et al. · arXiv · Sep 1, 2026

  Complex 3D spatial text to image generation requires models to convert natural language into stable visual geometry, not merely semantic appearance. Existing prompt-driven or layout-conditioned methods improve controllability, but often lac…

• Gaussian Core LoRA: Distribution-Aware Dynamic Adaptation for Broad Concept Erasure
  Qinghui Gong, Xunlei Chen, Yu-Xuan Zhang, Hua Meng et al. · arXiv · Sep 1, 2026

  Concept erasure aims to suppress unsafe, privacy-sensitive, or undesirable generations in text-to-image diffusion models while preserving benign semantics, visual quality, and deployment efficiency. Existing adapter-based methods, such as L…

• LISynSeg: Data-Centric Label-to-Image Synthesis for Cross-Modality Whole-Heart Segmentation
  Jiacheng Wang, Ivana Isgum, Ipek Oguz · arXiv · Aug 31, 2026

  Whole-heart segmentation (WHS) in computed tomography (CT) and magnetic resonance imaging (MRI) is affected by acquisition shifts and heterogeneous cardiac annotations. Existing WHS systems combine architectural design, transfer learning, a…

• Identity-Conditioned Latent Consistency Distillation for Face Synthesis
  Tiago Kienen Chaves, Bernardo Biesseck, David Menotti · arXiv · Aug 31, 2026

  Diffusion models have achieved strong results in high-fidelity image synthesis, but their iterative sampling process makes large-scale generation computationally expensive. This limitation is especially relevant when generating synthetic fa…

• GenEval 2: Addressing Benchmark Drift in Text-to-Image Evaluation
  Amita Kamath, Kai-Wei Chang, Ranjay Krishna, Luke S. Zettlemoyer et al. · arXiv.org · Dec 18, 2025

  Automating Text-to-Image (T2I) model evaluation is challenging; a judge model must be used to score correctness, and test prompts must be selected to be challenging for current T2I models but not the judge. We argue that satisfying these co…

• FLUX-Reason-6M & PRISM-Bench: A Million-Scale Text-to-Image Reasoning Dataset and Comprehensive Benchmark
  Rongyao Fang, Aldrich Yu, Chengqi Duan, Lin-Jiang Huang et al. · arXiv.org · Sep 11, 2025

  The advancement of open-source text-to-image (T2I) models has been hindered by the absence of large-scale, reasoning-focused datasets and comprehensive evaluation benchmarks, resulting in a performance gap compared to leading closed-source …

• Interleaving Reasoning for Better Text-to-Image Generation
  Wenxuan Huang, Shuang Chen, Zhe-Yong Xie, Shaosheng Cao et al. · arXiv.org · Sep 8, 2025

  Unified multimodal understanding and generation models recently have achieve significant improvement in image generation capability, yet a large gap remains in instruction following and detail preservation compared to systems that tightly c…