# TextFake: Benchmarking AI-Generated Image Detection on Text-Rich Images

arXiv: https://arxiv.org/abs/2606.01050 · HTML: https://mlsanity.com/paper/2606.01050/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 36 references that are in the map.
- First posted on arXiv 31 May 2026.

## Builds on (most cited first, 21 of 21)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [CNN-generated images are surprisingly easy to spot... for now](https://mlsanity.com/paper/1912.11035/) (2019)
- [Thinking in Frequency: Face Forgery Detection by Mining Frequency-aware Clues](https://mlsanity.com/paper/2007.09355/) (2020)
- [Towards Universal Fake Image Detectors that Generalize Across Generative Models](https://mlsanity.com/paper/2302.10174/) (2023)
- [GenImage: A Million-Scale Benchmark for Detecting AI-Generated Image](https://mlsanity.com/paper/2306.08571/) (2023)
- [Detecting and Simulating Artifacts in GAN Fake Images](https://mlsanity.com/paper/1907.06515/) (2019)
- [Watch your Up-Convolution: CNN Based Generative Deep Neural Networks are Failing to Reproduce Spectral Distributions](https://mlsanity.com/paper/2003.01826/) (2020)
- [The Creation and Detection of Deepfakes: A Survey](https://mlsanity.com/paper/2004.11138/) (2020)
- [Global Texture Enhancement for Fake Face Detection in the Wild](https://mlsanity.com/paper/2002.00133/) (2020)
- [PP-OCR: A Practical Ultra Lightweight OCR System](https://mlsanity.com/paper/2009.09941/) (2020)
- [ForgeryNet: A Versatile Benchmark for Comprehensive Forgery Analysis](https://mlsanity.com/paper/2103.05630/) (2021)
- [Improving Synthetic Image Detection Towards Generalization: An Image Transformation Perspective](https://mlsanity.com/paper/2408.06741/) (2024)
- [CIFAKE: Image Classification and Explainable Identification of AI-Generated Synthetic Images](https://mlsanity.com/paper/2303.14126/) (2023)
- [Any-Resolution AI-Generated Image Detection by Spectral Learning](https://mlsanity.com/paper/2411.19417/) (2024)
- [Real-Time Deepfake Detection in the Real-World](https://mlsanity.com/paper/2406.09398/) (2024)
- [Is Artificial Intelligence Generated Image Detection a Solved Problem?](https://mlsanity.com/paper/2505.12335/) (2025)
- [CO-SPY: Combining Semantic and Pixel Features to Detect Synthetic Images by AI](https://mlsanity.com/paper/2503.18286/) (2025)
- [Scaling Up AI-Generated Image Detection with Generator-Aware Prototypes](https://mlsanity.com/paper/2512.12982/) (2025)
- [SciFigDetect: A Benchmark for AI-Generated Scientific Figure Detection](https://mlsanity.com/paper/2604.08211/) (2026)
- [KIEval: Evaluation Metric for Document Key Information Extraction](https://mlsanity.com/paper/2503.05488/) (2025)

## Built on by (most cited first, 3 of 3)

- [A Benchmark Dataset for MLLM-Generated Image Detection: GPT Image2 & Nano Banana2](https://mlsanity.com/paper/2608.01258/) (2026)
- [Beyond Natural Images: Rethinking AI-Generated Image Detection in Documents](https://mlsanity.com/paper/2609.14352/) (2026)
- [False Claims, Credible Images: A Red-Teaming Benchmark for Commercial Image Generators](https://mlsanity.com/paper/2610.11112/) (2026)

## Abstract (arXiv)

Recent AI-generated image (AIGI) detectors perform well on natural-image benchmarks, but their behavior on text-rich forgeries, such as fabricated screenshots, documents, and news pages prevalent in misinformation, remains untested. We introduce TextFake, a 20,000-image benchmark for text-rich AIGI detection spanning 28 languages, 4 topic categories, and 2 scene modalities. Fake images are synthesized via a four-stage pipeline that annotates real images along three controlled dimensions and generates counterparts through distribution-aligned structured prompting, ruling out covariate shortcuts. Zero-shot evaluation of 14 specialized detectors and 3 frontier VLM APIs reveals a large systematic gap: no method exceeds 80% accuracy, with some dropping over 60% from natural-image benchmarks. Diagnostic evaluations identify three failure modes: the Text Density Curse, where dense glyphs overwhelm low-level detectors; Cloaking via Rendering Fidelity, where stronger text rendering suppresses enerative artifacts; and Threshold Collapse, where routine perturbations drive detectors toward chance-level performance.