xiaoooobai commited on
Commit
03bebe9
·
verified ·
1 Parent(s): cf473cb

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -1
README.md CHANGED
@@ -41,6 +41,10 @@ language:
41
  pretty_name: SVGenius
42
  size_categories:
43
  - n<1K
 
 
 
 
44
  ---
45
 
46
  # **SVGenius: Benchmarking LLMs in SVG Understanding, Editing and Generation**
@@ -54,4 +58,8 @@ size_categories:
54
  </a>
55
  <a href="https://zju-real.github.io/SVGenius/" target="_blank">
56
  <img alt="Webpage" src="https://img.shields.io/badge/%F0%9F%8C%8E_Website-SVGenius-green.svg" height="20" />
57
- </a>
 
 
 
 
 
41
  pretty_name: SVGenius
42
  size_categories:
43
  - n<1K
44
+ tags:
45
+ - svg-understanding
46
+ - svg-editing
47
+ - svg-generation
48
  ---
49
 
50
  # **SVGenius: Benchmarking LLMs in SVG Understanding, Editing and Generation**
 
58
  </a>
59
  <a href="https://zju-real.github.io/SVGenius/" target="_blank">
60
  <img alt="Webpage" src="https://img.shields.io/badge/%F0%9F%8C%8E_Website-SVGenius-green.svg" height="20" />
61
+ </a>
62
+
63
+ We introduce SVGenius, the first large-scale complexity-stratified benchmark accessing (M)LLMs capabilities across three progressive dimensions: **Understanding** (perceptua and semantic QA), **Editing** (bug fixing, code optimization, style editing), and **Generation** (text-to-SVG, image-to-SVG, style transfer). Built on real-world data from **24** application domains with systematic complexity stratification, SVGenius evaluates models through **8** task categories and **18** metrics. We assess **22** mainstream models spanning different scales, architectures, training paradigms, and accessibility levels.
64
+
65
+ <!-- <img src="docs/static/images/overview.jpg" width="100%"/> -->