# Investigating Text Insulation and Attention Mechanisms for Complex Visual Text Generation

arXiv: https://arxiv.org/abs/2503.23461 · HTML: https://mlsanity.com/paper/2503.23461/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- Builds on 42 of its 70 references that are in the map.
- First posted on arXiv 30 Mar 2025.

## Builds on (most cited first, 30 of 42)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [Pick-a-Pic: An Open Dataset of User Preferences for Text-to-Image Generation](https://mlsanity.com/paper/2305.01569/) (2023)
- [GenEval: An Object-Focused Framework for Evaluating Text-to-Image Alignment](https://mlsanity.com/paper/2310.11513/) (2023)
- [GLIGEN: Open-Set Grounded Text-to-Image Generation](https://mlsanity.com/paper/2301.07093/) (2023)
- [Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2301.13826/) (2023)
- [Improved Distribution Matching Distillation for Fast Image Synthesis](https://mlsanity.com/paper/2405.14867/) (2024)
- [Evaluating Text-to-Visual Generation with Image-to-Text Generation](https://mlsanity.com/paper/2404.01291/) (2024)
- [Training-Free Structured Diffusion Guidance for Compositional Text-to-Image Synthesis](https://mlsanity.com/paper/2212.05032/) (2022)
- [OpenVid-1M: A Large-Scale High-Quality Dataset for Text-to-video Generation](https://mlsanity.com/paper/2407.02371/) (2024)
- [Human Preference Score: Better Aligning Text-to-Image Models with Human Preference](https://mlsanity.com/paper/2303.14420/) (2023)
- [Simplifying, Stabilizing and Scaling Continuous-Time Consistency Models](https://mlsanity.com/paper/2410.11081/) (2024)
- [BoxDiff: Text-to-Image Synthesis with Training-Free Box-Constrained Diffusion](https://mlsanity.com/paper/2307.10816/) (2023)
- [TextDiffuser: Diffusion Models as Text Painters](https://mlsanity.com/paper/2305.10855/) (2023)
- [Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs](https://mlsanity.com/paper/2401.11708/) (2024)
- [AnyText: Multilingual Visual Text Generation And Editing](https://mlsanity.com/paper/2311.03054/) (2023)
- [TextDiffuser-2: Unleashing the Power of Language Models for Text Rendering](https://mlsanity.com/paper/2311.16465/) (2023)
- [GlyphControl: Glyph Conditional Control for Visual Text Generation](https://mlsanity.com/paper/2305.18259/) (2023)
- [MIGC: Multi-Instance Generation Controller for Text-to-Image Synthesis](https://mlsanity.com/paper/2402.05408/) (2024)
- [Scene Text Detection and Recognition: The Deep Learning Era](https://mlsanity.com/paper/1811.04256/) (2018)
- [Wavelet-based Fourier Information Interaction with Frequency Diffusion Adjustment for Underwater Image Restoration](https://mlsanity.com/paper/2311.16845/) (2023)

## Built on by (most cited first, 25 of 25)

- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer](https://mlsanity.com/paper/2511.22699/) (2025)
- [Cosmos 3: Omnimodal World Models for Physical AI](https://mlsanity.com/paper/2606.02800/) (2026)
- [Emu3.5: Native Multimodal Models are World Learners](https://mlsanity.com/paper/2510.26583/) (2025)
- [LongCat-Image Technical Report](https://mlsanity.com/paper/2512.07584/) (2025)
- [Unified Multimodal Understanding and Generation Models: Advances, Challenges, and Opportunities](https://mlsanity.com/paper/2505.02567/) (2025)
- [InternVL-U: Democratizing Unified Multimodal Models for Understanding, Reasoning, Generation and Editing](https://mlsanity.com/paper/2603.09877/) (2026)
- [Visual Text Processing: A Comprehensive Review and Unified Evaluation](https://mlsanity.com/paper/2504.21682/) (2025)
- [UltraHR-100K: Enhancing UHR Image Synthesis with A Large-Scale High-Quality Dataset](https://mlsanity.com/paper/2510.20661/) (2025)
- [DeepGen 1.0: A Lightweight Unified Multimodal Model for Advancing Image Generation and Editing](https://mlsanity.com/paper/2602.12205/) (2026)
- [ThinkGen: Generalized Thinking for Visual Generation](https://mlsanity.com/paper/2512.23568/) (2025)
- [Ovis-Image Technical Report](https://mlsanity.com/paper/2511.22982/) (2025)
- [Training-Free Text-Guided Color Editing with Multi-Modal Diffusion Transformer](https://mlsanity.com/paper/2508.09131/) (2025)
- [FreeText: Training-Free Text Rendering in Diffusion Transformers via Attention Localization and Spectral Glyph Injection](https://mlsanity.com/paper/2601.00535/) (2026)
- [MIND-Edit: MLLM Insight-Driven Editing via Language-Vision Projection](https://mlsanity.com/paper/2505.19149/) (2025)
- [GlyphBanana: Advancing Precise Text Rendering Through Agentic Workflows](https://mlsanity.com/paper/2603.12155/) (2026)
- [LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes](https://mlsanity.com/paper/2609.03796/) (2026)
- [TextGuider: Training-Free Guidance for Text Rendering via Attention Alignment](https://mlsanity.com/paper/2512.09350/) (2025)
- [VAR RL Done Right: Tackling Asynchronous Policy Conflicts in Visual Autoregressive Generation](https://mlsanity.com/paper/2601.02256/) (2026)
- [i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models](https://mlsanity.com/paper/2606.11289/) (2026)
- [GlyphAnchor: Enhancing Visual Text Rendering via Position-Anchored Glyph Priors](https://mlsanity.com/paper/2609.02349/) (2026)
- [Amber-Image: Efficient Compression of Large-Scale Diffusion Transformers](https://mlsanity.com/paper/2602.17047/) (2026)
- [SeFi-Image: A Text-to-Image Foundation Model with Semantic-First Diffusion](https://mlsanity.com/paper/2606.22568/) (2026)
- [S1-Omni: A Unified Multimodal Reasoning Model for Scientific Understanding, Prediction, and Generation](https://mlsanity.com/paper/2607.15686/) (2026)
- [TextVBench: Evaluating Text Rendering and Preservation in Video Generation](https://mlsanity.com/paper/2610.13412/) (2026)

## Abstract (arXiv)

In this paper, we present TextCrafter, a Complex Visual Text Generation (CVTG) framework inspired by selective visual attention in cognitive science, and introduce the "Text Insulation-and-Attention" mechanisms. To implement the selective-attention principle that selection operates on discrete objects, we propose a novel Bottleneck-aware Constrained Reinforcement Learning for Multi-text Insulation, which substantially improves text-rendering performance on the strong Qwen-Image pretrained model without introducing additional parameters. To align with the selective concentration principle in human vision, we introduce a text-oriented attention module with a novel Quotation-guided Attention Gate that further improves generation quality for each text instance. Our Reinforcement Learning based text insulation approach attains state-of-the-art results, and incorporating text-oriented attention yields additional gains on top of an already strong baseline. More importantly, we introduce CVTG-2K, a benchmark comprising 2,000 complex visual-text prompts. These prompts vary in positions, quantities, lengths, and attributes, and span diverse real-world scenarios. Extensive evaluations on CVTG-2K, CVTG-Hard, LongText-Bench, and Geneval datasets confirm the effectiveness of TextCrafter. Despite using substantially fewer resources (i.e., 4 GPUs) than industrial-scale models (e.g., Qwen-Image, GPT Image, and Seedream), TextCrafter achieves superior performance in mitigating text misgeneration, omissions, and hallucinations.