Skip to content

Commit 6faaf28

Browse files
yaoyifan-yyfyaoyifan-yyf
authored
Create index.html
1 parent 6870e4b commit 6faaf28

File tree

1 file changed

+304
-0
lines changed

1 file changed

+304
-0
lines changed

‎index.html‎

Lines changed: 304 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,304 @@
1+
<!DOCTYPE html>
2+
<html lang="zh-CN">
3+
<head>
4+
<meta charset="UTF-8">
5+
<meta name="viewport" content="width=device-width, initial-scale=1.0">
6+
<title>Falcon: Text-to-SQL Benchmark</title>
7+
<meta name="description" content="Falcon is a continuously-evolving, high-quality benchmark for natural-language-to-SQL (Text2SQL) systems.">
8+
<link href="https://fonts.googleapis.com/css2?family=Inter:wght@300;400;600;700&display=swap" rel="stylesheet">
9+
<style>
10+
:root {
11+
--primary: #0969da;
12+
--dark: #1f2328;
13+
--light: #f6f8fa;
14+
--border: #d0d7de;
15+
}
16+
17+
body {
18+
font-family: 'Inter', -apple-system, BlinkMacSystemFont, sans-serif;
19+
line-height: 1.6;
20+
color: var(--dark);
21+
margin: 0;
22+
padding: 0;
23+
}
24+
25+
/* 导航栏 */
26+
.navbar {
27+
border-bottom: 1px solid var(--border);
28+
padding: 1rem 0;
29+
position: sticky;
30+
top: 0;
31+
background: white;
32+
z-index: 100;
33+
}
34+
35+
.container {
36+
max-width: 1000px;
37+
margin: 0 auto;
38+
padding: 0 20px;
39+
}
40+
41+
.nav-content {
42+
display: flex;
43+
justify-content: space-between;
44+
align-items: center;
45+
}
46+
47+
.logo {
48+
font-weight: 700;
49+
font-size: 1.25rem;
50+
text-decoration: none;
51+
color: var(--dark);
52+
display: flex;
53+
align-items: center;
54+
gap: 10px;
55+
}
56+
57+
.nav-links a {
58+
margin-left: 20px;
59+
text-decoration: none;
60+
color: #57606a;
61+
font-size: 0.95rem;
62+
}
63+
.nav-links a:hover { color: var(--primary); }
64+
65+
/* Hero 区域 */
66+
.hero {
67+
background-color: var(--light);
68+
padding: 4rem 0;
69+
text-align: center;
70+
border-bottom: 1px solid var(--border);
71+
}
72+
73+
.hero h1 {
74+
font-size: 2.5rem;
75+
margin-bottom: 1rem;
76+
color: var(--dark);
77+
}
78+
79+
.hero p {
80+
font-size: 1.25rem;
81+
color: #57606a;
82+
max-width: 700px;
83+
margin: 0 auto 2rem;
84+
}
85+
86+
.btn-group {
87+
display: flex;
88+
gap: 15px;
89+
justify-content: center;
90+
}
91+
92+
.btn {
93+
padding: 10px 20px;
94+
border-radius: 6px;
95+
text-decoration: none;
96+
font-weight: 600;
97+
transition: all 0.2s;
98+
}
99+
100+
.btn-primary {
101+
background-color: var(--primary);
102+
color: white;
103+
}
104+
.btn-primary:hover { background-color: #0856b3; }
105+
106+
.btn-outline {
107+
border: 1px solid var(--border);
108+
background-color: white;
109+
color: var(--dark);
110+
}
111+
.btn-outline:hover { background-color: #f3f4f6; }
112+
113+
/* 内容区块 */
114+
section { padding: 3rem 0; }
115+
116+
h2 {
117+
border-bottom: 1px solid var(--border);
118+
padding-bottom: 0.5rem;
119+
margin-bottom: 1.5rem;
120+
}
121+
122+
.features-grid {
123+
display: grid;
124+
grid-template-columns: repeat(auto-fit, minmax(300px, 1fr));
125+
gap: 2rem;
126+
margin-top: 2rem;
127+
}
128+
129+
.feature-card {
130+
background: white;
131+
border: 1px solid var(--border);
132+
border-radius: 6px;
133+
padding: 1.5rem;
134+
}
135+
136+
.feature-card h3 { margin-top: 0; color: var(--primary); }
137+
138+
/* 排行榜表格 */
139+
table {
140+
width: 100%;
141+
border-collapse: collapse;
142+
margin-top: 1rem;
143+
font-size: 0.95rem;
144+
}
145+
146+
th, td {
147+
text-align: left;
148+
padding: 12px;
149+
border-bottom: 1px solid var(--border);
150+
}
151+
152+
th { background-color: var(--light); font-weight: 600; }
153+
154+
/* 代码块 */
155+
pre {
156+
background: #f6f8fa;
157+
padding: 1rem;
158+
border-radius: 6px;
159+
overflow-x: auto;
160+
font-size: 0.9rem;
161+
}
162+
163+
footer {
164+
text-align: center;
165+
padding: 2rem 0;
166+
color: #57606a;
167+
font-size: 0.9rem;
168+
border-top: 1px solid var(--border);
169+
margin-top: 2rem;
170+
}
171+
172+
@media (max-width: 600px) {
173+
.nav-links { display: none; }
174+
.hero h1 { font-size: 2rem; }
175+
table { display: block; overflow-x: auto; }
176+
}
177+
</style>
178+
</head>
179+
<body>
180+
181+
<nav class="navbar">
182+
<div class="container nav-content">
183+
<a href="#" class="logo">🦅 Falcon Benchmark</a>
184+
<div class="nav-links">
185+
<a href="#about">About</a>
186+
<a href="#features">Features</a>
187+
<a href="#leaderboard">Leaderboard</a>
188+
<a href="#citation">Citation</a>
189+
<a href="https://github.com/eosphoros-ai/Falcon" target="_blank">GitHub</a>
190+
</div>
191+
</div>
192+
</nav>
193+
194+
<div class="hero">
195+
<div class="container">
196+
<h1>Falcon: Text-to-SQL Benchmark</h1>
197+
<p>A continuously-evolving, high-quality benchmark designed to stress-test models under complex, cross-domain analytical scenarios.</p>
198+
<div class="btn-group">
199+
<a href="https://github.com/eosphoros-ai/Falcon" class="btn btn-primary">View on GitHub</a>
200+
<a href="https://arxiv.org/abs/2510.24762" class="btn btn-outline">Read Paper</a>
201+
</div>
202+
</div>
203+
</div>
204+
205+
<div class="container">
206+
<section id="about">
207+
<h2>About Falcon</h2>
208+
<p>
209+
Falcon is built on public Kaggle datasets covering <strong>Finance, Internet, and Retail</strong> domains.
210+
Each domain is paired with question sets of incremental difficulty levels so that model capability can be measured more finely.
211+
</p>
212+
<p>
213+
The current release includes 500 Chinese questions derived from Kaggle datasets, with more questions focusing on fuzzy linguistic expressions and real Ant Group production scenarios in preparation.
214+
</p>
215+
</section>
216+
217+
<section id="features">
218+
<h2>Key Challenges</h2>
219+
<div class="features-grid">
220+
<div class="feature-card">
221+
<h3>SQL-Computational Difficulties</h3>
222+
<p>Designed to test complex logic including:</p>
223+
<ul>
224+
<li>Multi-table joins</li>
225+
<li>Nested CTEs (Common Table Expressions)</li>
226+
<li>Window functions & Ranking</li>
227+
<li>Type casting & Regular-expression filters</li>
228+
</ul>
229+
</div>
230+
<div class="feature-card">
231+
<h3>Linguistic Difficulties</h3>
232+
<p>Focuses on natural language ambiguity:</p>
233+
<ul>
234+
<li>Chinese fuzzy time expressions</li>
235+
<li>Colloquial business jargon</li>
236+
<li>Ellipsis (omitted information)</li>
237+
<li>Multi-intent questions</li>
238+
</ul>
239+
</div>
240+
</div>
241+
</section>
242+
243+
<section id="leaderboard">
244+
<h2>🏆 Leaderboard</h2>
245+
<p><em>Note: This is a placeholder. Please update with real evaluation results.</em></p>
246+
<table>
247+
<thead>
248+
<tr>
249+
<th>Rank</th>
250+
<th>Model</th>
251+
<th>Method</th>
252+
<th>Execution Accuracy</th>
253+
<th>Date</th>
254+
</tr>
255+
</thead>
256+
<tbody>
257+
<tr>
258+
<td>1</td>
259+
<td>GPT-4o</td>
260+
<td>Few-shot CoT</td>
261+
<td>--</td>
262+
<td>2024-05</td>
263+
</tr>
264+
<tr>
265+
<td>2</td>
266+
<td>Claude 3.5 Sonnet</td>
267+
<td>Zero-shot</td>
268+
<td>--</td>
269+
<td>2024-06</td>
270+
</tr>
271+
<tr>
272+
<td>3</td>
273+
<td>Llama-3-70B</td>
274+
<td>Fine-tuned</td>
275+
<td>--</td>
276+
<td>2024-04</td>
277+
</tr>
278+
</tbody>
279+
</table>
280+
</section>
281+
282+
<section id="citation">
283+
<h2>📝 Citation</h2>
284+
<p>For methodology, dataset construction details, and evaluation protocols, please refer to our arXiv paper:</p>
285+
<pre><code>@article{falcon2025,
286+
title={Falcon: A New Text-to-SQL Benchmark},
287+
author={Eosphoros AI Team},
288+
journal={arXiv preprint arXiv:2510.24762},
289+
year={2025}
290+
}</code></pre>
291+
</section>
292+
</div>
293+
294+
<footer>
295+
<div class="container">
296+
<p>
297+
Licensed under Apache-2.0. <br>
298+
Maintained by <a href="https://github.com/eosphoros-ai">eosphoros-ai</a>.
299+
</p>
300+
</div>
301+
</footer>
302+
303+
</body>
304+
</html>

0 commit comments

Comments
 (0)