VecGlypher: Unified Vector Glyph Generation with Language Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908851981254656 |
|---|---|
| author | Huang, Xiaoke Gauri, Bhavul Ng, Kam Woh Ng, Tony Xu, Mengmeng Liu, Zhiheng Ren, Weiming An, Zhaochong Zhou, Zijian Qiu, Haonan Zhou, Yuyin He, Sen Wang, Ziheng Xiang, Tao Han, Xiao |
| author_facet | Huang, Xiaoke Gauri, Bhavul Ng, Kam Woh Ng, Tony Xu, Mengmeng Liu, Zhiheng Ren, Weiming An, Zhaochong Zhou, Zijian Qiu, Haonan Zhou, Yuyin He, Sen Wang, Ziheng Xiang, Tao Han, Xiao |
| contents | Vector glyphs are the atomic units of digital typography, yet most learning-based pipelines still depend on carefully curated exemplar sheets and raster-to-vector postprocessing, which limits accessibility and editability. We introduce VecGlypher, a single multimodal language model that generates high-fidelity vector glyphs directly from text descriptions or image exemplars. Given a style prompt, optional reference glyph images, and a target character, VecGlypher autoregressively emits SVG path tokens, avoiding raster intermediates and producing editable, watertight outlines in one pass. A typography-aware data and training recipe makes this possible: (i) a large-scale continuation stage on 39K noisy Envato fonts to master SVG syntax and long-horizon geometry, followed by (ii) post-training on 2.5K expert-annotated Google Fonts with descriptive tags and exemplars to align language and imagery with geometry; preprocessing normalizes coordinate frames, canonicalizes paths, de-duplicates families, and quantizes coordinates for stable long-sequence decoding. On cross-family OOD evaluation, VecGlypher substantially outperforms both general-purpose LLMs and specialized vector-font baselines for text-only generation, while image-referenced generation reaches a state-of-the-art performance, with marked gains over DeepVecFont-v2 and DualVector. Ablations show that model scale and the two-stage recipe are critical and that absolute-coordinate serialization yields the best geometry. VecGlypher lowers the barrier to font creation by letting users design with words or exemplars, and provides a scalable foundation for future multimodal design tools. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_21461 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | VecGlypher: Unified Vector Glyph Generation with Language Models Huang, Xiaoke Gauri, Bhavul Ng, Kam Woh Ng, Tony Xu, Mengmeng Liu, Zhiheng Ren, Weiming An, Zhaochong Zhou, Zijian Qiu, Haonan Zhou, Yuyin He, Sen Wang, Ziheng Xiang, Tao Han, Xiao Computation and Language Vector glyphs are the atomic units of digital typography, yet most learning-based pipelines still depend on carefully curated exemplar sheets and raster-to-vector postprocessing, which limits accessibility and editability. We introduce VecGlypher, a single multimodal language model that generates high-fidelity vector glyphs directly from text descriptions or image exemplars. Given a style prompt, optional reference glyph images, and a target character, VecGlypher autoregressively emits SVG path tokens, avoiding raster intermediates and producing editable, watertight outlines in one pass. A typography-aware data and training recipe makes this possible: (i) a large-scale continuation stage on 39K noisy Envato fonts to master SVG syntax and long-horizon geometry, followed by (ii) post-training on 2.5K expert-annotated Google Fonts with descriptive tags and exemplars to align language and imagery with geometry; preprocessing normalizes coordinate frames, canonicalizes paths, de-duplicates families, and quantizes coordinates for stable long-sequence decoding. On cross-family OOD evaluation, VecGlypher substantially outperforms both general-purpose LLMs and specialized vector-font baselines for text-only generation, while image-referenced generation reaches a state-of-the-art performance, with marked gains over DeepVecFont-v2 and DualVector. Ablations show that model scale and the two-stage recipe are critical and that absolute-coordinate serialization yields the best geometry. VecGlypher lowers the barrier to font creation by letting users design with words or exemplars, and provides a scalable foundation for future multimodal design tools. |
| title | VecGlypher: Unified Vector Glyph Generation with Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2602.21461 |