PixText fixes text in generated images: 83% glyph pass rate vs 37% for baselines
ATH-MaaS · hf · 2026-10-15
Researchers from Zhejiang University and Alibaba introduce PixText, which formulates scene-text refinement as grouped image generation in pixel space: conditioned on the original image, it generates only target text regions, preserves unedited content, and uses MeanFlow for four-step generation — addressing stroke loss in VAE-based methods and background shifts in full-image editing.
The team also releases the Text Refinement Bench. On it, PixText achieves 83.06% glyph pass rate and 94.66% exact target-character OCR accuracy, versus 36.77% and 52.31% for the strongest baseline. The four-step MeanFlow variant keeps an 81.84% pass rate at under one amortized GPU-second per image.
More from Multimodal
- Fish Audio's Drama-3 voice-cloning Space trends on Hugging Face — fishaudio · 2026-10-15
- Midjourney v8.2 style test: bold flat-ink illustration looks, sref code included — azed_ai · 2026-10-15
- Sarvam AI launches Bulbul V4 Flash, a faster compact TTS model for Indian languages — CurieuxExplorer · 2026-10-15
- Maya-2 debuts as open-weight Indic voice model, ranks #2 on Hindi TTS leaderboard — CurieuxExplorer · 2026-10-15
- TextVerifier catches flawed Chinese glyph rendering with F1 0.844, far above OCR and VLMs — ATH-MaaS · 2026-10-15
- AI Video Creators: 'Amazing Visuals Are Solved — Unique IP and Characters Are What's Left' — AIandDesign · 2026-10-15