# MULTITEXTEDIT: Benchmarking Cross-Lingual Degradation in Text-in-Image Editing

arXiv: https://arxiv.org/abs/2605.08163 · HTML: https://mlsanity.com/paper/2605.08163/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 33 references that are in the map.
- First posted on arXiv 4 May 2026.

## Builds on (most cited first, 22 of 22)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [InstructPix2Pix: Learning to Follow Image Editing Instructions](https://mlsanity.com/paper/2211.09800/) (2022)
- [Emerging Properties in Unified Multimodal Pretraining](https://mlsanity.com/paper/2505.14683/) (2025)
- [An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition](https://mlsanity.com/paper/1507.05717/) (2015)
- [TextDiffuser: Diffusion Models as Text Painters](https://mlsanity.com/paper/2305.10855/) (2023)
- [What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis](https://mlsanity.com/paper/1904.01906/) (2019)
- [AnyText: Multilingual Visual Text Generation And Editing](https://mlsanity.com/paper/2311.03054/) (2023)
- [Scene Text Recognition with Permuted Autoregressive Sequence Models](https://mlsanity.com/paper/2207.06966/) (2022)
- [TextDiffuser-2: Unleashing the Power of Language Models for Text Rendering](https://mlsanity.com/paper/2311.16465/) (2023)
- [KRIS-Bench: Benchmarking Next-Level Intelligent Image Editing Models](https://mlsanity.com/paper/2505.16707/) (2025)
- [AnyText2: Visual Text Generation and Editing With Customizable Attributes](https://mlsanity.com/paper/2411.15245/) (2024)
- [FLUX-Text: A Simple and Advanced Diffusion Transformer Baseline for Scene Text Editing](https://mlsanity.com/paper/2505.03329/) (2025)
- [TextCtrl: Diffusion-based Scene Text Editing with Prior Guidance Control](https://mlsanity.com/paper/2410.10133/) (2024)
- [FireRed-Image-Edit-1.0 Technical Report](https://mlsanity.com/paper/2602.13344/) (2026)
- [Visual Text Processing: A Comprehensive Review and Unified Evaluation](https://mlsanity.com/paper/2504.21682/) (2025)
- [GIE-Bench: Towards Grounded Evaluation for Text-Guided Image Editing](https://mlsanity.com/paper/2505.11493/) (2025)
- [Choose What You Need: Disentangled Representation Learning for Scene Text Recognition, Removal and Editing](https://mlsanity.com/paper/2405.04377/) (2024)
- [GlyphMastero: A Glyph Encoder for High-Fidelity Scene Text Editing](https://mlsanity.com/paper/2505.04915/) (2025)
- [TextEditBench: Evaluating Reasoning-aware Text Editing Beyond Rendering](https://mlsanity.com/paper/2512.16270/) (2025)
- [TextMastero: Mastering High-Quality Scene Text Editing in Diverse Languages and Styles](https://mlsanity.com/paper/2408.10623/) (2024)
- [IMTBench: A Multi-Scenario Cross-Modal Collaborative Evaluation Benchmark for In-Image Machine Translation](https://mlsanity.com/paper/2603.10495/) (2026)

## Built on by (most cited first, 1 of 1)

- [Efficient Scene-Text Refinement via Pixel-Space Grouped Image Generation](https://mlsanity.com/paper/2610.18284/) (2026)

## Abstract (arXiv)

Text-in-image editing has become a key capability for visual content creation, yet existing benchmarks remain overwhelmingly English-centric and often conflate visual plausibility with semantic correctness. We introduce MULTITEXTEDIT, a controlled benchmark of 3,600 instances spanning 12 typologically diverse languages, 5 visual domains, and 7 editing operations. Language variants of each instance share a common visual base and are paired with a human-edited reference and region masks, isolating the language variable for cross-lingual comparison. To capture script-level errors that coarse text-matching metrics miss, such as missing diacritics, reversed RTL order, and mixed-script renderings, we introduce a language fidelity (LSF) metric scored by a two-stage LVM protocol that first traces the edited target text and then judges it in isolation, reaching a quadratic-weighted kappa of 0.76 against native-speaker annotators. Evaluating 12 open-source and proprietary systems with LSF alongside standard semantic and mask-aware pixel metrics, we find pronounced cross-lingual degradation for every model, largest on Hebrew and Arabic and smallest on Dutch and Spanish, and concentrated in text accuracy and script fidelity rather than in coarse structural dimensions. We also uncover a pervasive semantic and pixel mismatch, where outputs preserve global layout and background fidelity yet distort script-specific forms.