# Beyond Legibility: Benchmarking Visual Text Rendering and In-Place Editing in Unified Video Generation

arXiv: https://arxiv.org/abs/2609.36598 · HTML: https://mlsanity.com/paper/2609.36598/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 36 of its 46 references that are in the map.
- First posted on arXiv 29 Sep 2026.

## Builds on (most cited first, 30 of 36)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [Diffusion Model Alignment Using Direct Preference Optimization](https://mlsanity.com/paper/2311.12908/) (2023)
- [Synthetic Data for Text Localisation in Natural Images](https://mlsanity.com/paper/1604.06646/) (2016)
- [EvalCrafter: Benchmarking and Evaluating Large Video Generation Models](https://mlsanity.com/paper/2310.11440/) (2023)
- [VBench-2.0: Advancing Video Generation Benchmark Suite for Intrinsic Faithfulness](https://mlsanity.com/paper/2503.21755/) (2025)
- [LTX-2: Efficient Joint Audio-Visual Foundation Model](https://mlsanity.com/paper/2601.03233/) (2026)
- [TextDiffuser: Diffusion Models as Text Painters](https://mlsanity.com/paper/2305.10855/) (2023)
- [T2V-CompBench: A Comprehensive Benchmark for Compositional Text-to-video Generation](https://mlsanity.com/paper/2407.14505/) (2024)
- [TextOCR: Towards large-scale end-to-end reasoning for arbitrary-shaped scene text](https://mlsanity.com/paper/2105.05486/) (2021)
- [AnyText: Multilingual Visual Text Generation And Editing](https://mlsanity.com/paper/2311.03054/) (2023)
- [HunyuanVideo 1.5 Technical Report](https://mlsanity.com/paper/2511.18870/) (2025)
- [GlyphControl: Glyph Conditional Control for Visual Text Generation](https://mlsanity.com/paper/2305.18259/) (2023)
- [ICDAR 2019 Competition on Large-scale Street View Text with Partial Labeling -- RRC-LSVT](https://mlsanity.com/paper/1909.07741/) (2019)
- [Glyph-ByT5: A Customized Text Encoder for Accurate Visual Text Rendering](https://mlsanity.com/paper/2403.09622/) (2024)
- [Video-Bench: Human-Aligned Video Generation Benchmark](https://mlsanity.com/paper/2504.04907/) (2025)
- [TextStyleBrush: Transfer of Text Aesthetics from a Single Example](https://mlsanity.com/paper/2106.08385/) (2021)
- [TextCtrl: Diffusion-based Scene Text Editing with Prior Guidance Control](https://mlsanity.com/paper/2410.10133/) (2024)
- [UnrealText: Synthesizing Realistic Scene Text Images from the Unreal World](https://mlsanity.com/paper/2003.10608/) (2020)
- [Kiwi-Edit: Versatile Video Editing via Instruction and Reference Guidance](https://mlsanity.com/paper/2603.02175/) (2026)
- [VABench: A Comprehensive Benchmark for Audio-Video Generation](https://mlsanity.com/paper/2512.09299/) (2025)
- [EasyText: Controllable Diffusion Transformer for Multilingual Text Rendering](https://mlsanity.com/paper/2505.24417/) (2025)
- [Visual Text Processing: A Comprehensive Review and Unified Evaluation](https://mlsanity.com/paper/2504.21682/) (2025)
- [Don't Forget Me: Accurate Background Recovery for Text Removal via Modeling Local-Global Context](https://mlsanity.com/paper/2207.10273/) (2022)
- [TextPecker: Rewarding Structural Anomaly Quantification for Enhancing Visual Text Rendering](https://mlsanity.com/paper/2602.20903/) (2026)
- [MSAVBench: Towards Comprehensive and Reliable Evaluation of Multi-Shot Audio-Video Generation](https://mlsanity.com/paper/2605.20183/) (2026)

## Abstract (arXiv)

A video can exhibit convincing motion and photorealism yet fail immediately when visual text collapses. Unlike generic scene content, visual text is unforgiving in video generation: minor stroke corruption, temporal instability, or editing errors instantly break legibility and realism. Existing benchmarks overlook this challenge by treating text as incidental or using static OCR metrics that ignore temporal dynamics. We introduce VidScribe, a unified diagnostic benchmark spanning four generation regimes: writing from language (T2V), transferring text identity from a reference (R2V), sustaining text under dynamics (I2V), and localized text editing (V2V). VidScribe contains 803 human-verified samples across a 12-axis conditionally orthogonal factor space covering Intrinsic Text Properties, Physical Imaging Conditions, and Temporal Behavior. For reliable evaluation, we build a track-grounded, gated suite with 11 shared metrics and 2 task-specific probes under strict measurability conditions. Benchmarking 11 commercial and open-source systems shows that video text capability is non-monolithic, with content recognition decoupled from stroke-level glyph correctness. Performance is highly task-asymmetric: I2V sustains text most reliably, whereas V2V editing is the primary bottleneck. Counter-intuitively, degradation concentrates on a small subset of text-centric structural and temporal factors rather than adverse imaging conditions. Further probes show that visual references improve glyph and typographic fidelity rather than content accuracy, while localized editing fails to isolate target text without corrupting undeclared source text. Beyond evaluation, VidScribe also provides an actionable training signal, where benchmark-aligned preference optimization measurably improves visual text generation. https://huggingface.co/datasets/Vicky0720/VidScribe.