Skip to content

Commit cd751ea

Browse files
wenhuchenwenhuchen
committed
initial web page
1 parent 09050f4 commit cd751ea

22 files changed

+29947
-0
lines changed

‎index.html‎

Lines changed: 159 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,159 @@
1+
<!DOCTYPE html>
2+
<html>
3+
<head>
4+
<meta charset="utf-8">
5+
<meta name="description" content="AceCoder: Acing Coder RL via Automated Test-Case Synthesis">
6+
<meta property="og:title" content="AceCoder: Acing Coder RL via Automated Test-Case Synthesis" />
7+
<meta property="og:description" content="We propose Critique Fine-Tuning (CFT), where models learn to critique noisy responses rather than simply imitate correct ones." />
8+
<meta property="og:url" content="https://github.com/TIGER-AI-Lab/AceCoder" />
9+
<meta property="og:image" content="" />
10+
<meta property="og:image:width" content="1200" />
11+
<meta property="og:image:height" content="630" />
12+
13+
<title>Critique Fine-Tuning</title>
14+
<link href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro" rel="stylesheet">
15+
<link rel="stylesheet" href="static/css/bulma.min.css">
16+
<link rel="stylesheet" href="https://cdn.jsdelivr.net/gh/jpswalsh/academicons@1/css/academicons.min.css">
17+
<link rel="stylesheet" href="static/css/index.css">
18+
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.2/css/all.min.css">
19+
</head>
20+
21+
<body>
22+
<section class="hero">
23+
<div class="hero-body">
24+
<div class="container is-max-desktop">
25+
<div class="columns is-centered">
26+
<div class="column has-text-centered">
27+
<h1 class="title is-1 publication-title">
28+
AceCoder: Acing Coder RL via Automated Test-Case Synthesis
29+
</h1>
30+
<div class="is-size-5 publication-authors">
31+
<span class="author-block">
32+
<sup>♠️</sup>Huaye Zeng,
33+
</span>
34+
<span class="author-block">
35+
<sup>♠️</sup>Dongfu Jiang,
36+
</span>
37+
<span class="author-block">
38+
Haozhe Wang
39+
</span>
40+
<span class="author-block">
41+
Ping Nie
42+
</span>
43+
<span class="author-block">
44+
Xiaotong Chen
45+
</span>
46+
<span class="author-block">
47+
<sup>♠️</sup>Wenhu Chen
48+
</span>
49+
</div>
50+
51+
<div class="column has-text-centered">
52+
<div class="publication-links">
53+
<!-- GitHub link -->
54+
<span class="link-block">
55+
<a href="https://github.com/TIGER-AI-Lab/AceCoder" target="_blank"
56+
class="external-link button is-normal is-rounded is-dark">
57+
<span class="icon">
58+
<i class="fab fa-github"></i>
59+
</span>
60+
<span>Code</span>
61+
</a>
62+
</span>
63+
64+
<!-- Paper link (placeholder) -->
65+
<span class="link-block">
66+
<a href="#" target="_blank"
67+
class="external-link button is-normal is-rounded is-dark">
68+
<span class="icon">
69+
<i class="ai ai-arxiv"></i>
70+
</span>
71+
<span>Paper</span>
72+
</a>
73+
</span>
74+
75+
<span class="link-block">
76+
<a href="#" target="_blank"
77+
class="external-link button is-normal is-rounded is-dark">
78+
<span class="icon">
79+
🤗
80+
</span>
81+
<span>Dataset</span>
82+
</a>
83+
</span>
84+
85+
<span class="link-block">
86+
<a href="#" target="_blank"
87+
class="external-link button is-normal is-rounded is-dark">
88+
<span class="icon">
89+
🤗
90+
</span>
91+
<span>Models</span>
92+
</a>
93+
</span>
94+
</div>
95+
</div>
96+
97+
<div class="is-size-5 publication-authors">
98+
<span class="author-block">
99+
<sup>♠️</sup>University of Waterloo,
100+
HKUST,
101+
NetMind.AI
102+
</span>
103+
<br>
104+
<span class="author-block">
105+
<small>
106+
[email protected],
107+
[email protected]
108+
</small>
109+
</span>
110+
</div>
111+
112+
</div>
113+
</div>
114+
</div>
115+
</div>
116+
</section>
117+
118+
<section class="section hero is-light">
119+
<div class="container is-max-desktop">
120+
<div class="columns is-centered has-text-centered">
121+
<div class="column is-four-fifths">
122+
<h1 class="title is-3">Abstract</h1>
123+
<div class="content has-text-justified">
124+
<p>
125+
Recent years have witnessed the great performance of code model in code generation, code fix, etc. However, most recent work has been focused on supervised-fine-tuning (SFT) while the potential of reinforcement learning in training code models has been untapped. This is mostly hindered by the lack of reliable reward signals in the code domain. In this paper, we aim to empower code model training with automated test-case synthesis on a large scale. Specifically, we design a pipeline to synthesize large-scale (question, test-cases) pairs from existing seed code data.
126+
</p>
127+
</div>
128+
</div>
129+
</div>
130+
</div>
131+
</section>
132+
133+
<!-- BibTeX citation -->
134+
<section class="section" id="BibTeX">
135+
<div class="container is-max-desktop content">
136+
<h2 class="title">Reference</h2>
137+
Please kindly cite our paper if you use our code or results:
138+
<pre><code>
139+
140+
</code></pre>
141+
</div>
142+
</section>
143+
144+
<footer class="footer">
145+
<div class="container">
146+
<div class="columns is-centered">
147+
<div class="column is-8">
148+
<div class="content has-text-centered">
149+
<p>
150+
This website is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/" target="_blank">Creative Commons Attribution-ShareAlike 4.0 International License</a>.
151+
</p>
152+
</div>
153+
</div>
154+
</div>
155+
</div>
156+
</footer>
157+
158+
</body>
159+
</html>

‎static/css/bulma-carousel.min.css‎

Lines changed: 1 addition & 0 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

0 commit comments

Comments
 (0)