-
Notifications
You must be signed in to change notification settings - Fork 3
Expand file tree
/
Copy pathindex.html
More file actions
275 lines (238 loc) · 10.7 KB
/
index.html
File metadata and controls
275 lines (238 loc) · 10.7 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="description" content="GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning.">
<meta name="keywords" content="Large Language Model, Test-Time Scaling, Process Reward Model">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning</title>
<!-- Global site tag (gtag.js) - Google Analytics -->
<script async src="https://www.googletagmanager.com/gtag/js?id=G-PYVRSFMDRL"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag() {
dataLayer.push(arguments);
}
gtag('js', new Date());
gtag('config', 'G-PYVRSFMDRL');
</script>
<link href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro" rel="stylesheet">
<link rel="stylesheet" href="./static/css/bulma.min.css">
<link rel="stylesheet" href="./static/css/bulma-carousel.min.css">
<link rel="stylesheet" href="./static/css/bulma-slider.min.css">
<link rel="stylesheet" href="./static/css/fontawesome.all.min.css">
<link rel="stylesheet"
href="https://cdn.jsdelivr.net/gh/jpswalsh/academicons@1/css/academicons.min.css">
<link rel="stylesheet" href="./static/css/index.css">
<!-- <link rel="icon" href="./static/images/favicon.svg"> -->
<script src="https://ajax.googleapis.com/ajax/libs/jquery/3.5.1/jquery.min.js"></script>
<script defer src="./static/js/fontawesome.all.min.js"></script>
<script src="./static/js/bulma-carousel.min.js"></script>
<script src="./static/js/bulma-slider.min.js"></script>
<script src="./static/js/index.js"></script>
</head>
<body>
<section class="hero">
<div class="hero-body">
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column has-text-centered">
<!-- Title -->
<h1 class="title is-1 publication-title is-bold">
GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning
</h1>
<!-- Author list -->
<div class="is-size-5 publication-authors">
<span class="author-block">
<a href="https://openreview.net/profile?id=~Jian_Zhao12">Jian Zhao</a><sup>1,3*</sup>,
</span>
<span class="author-block">
<a href="https://ryanliu112.github.io">Runze Liu</a><sup>1,2*</sup>,
</span>
<span class="author-block">
<a href="https://iseesaw.github.io">Kaiyan Zhang</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="https://openreview.net/profile?id=~Zhimu_Zhou2">Zhimu Zhou</a><sup>3</sup>,
</span>
<span class="author-block">
<a href="https://scholar.google.com/citations?user=mWpsZ1wAAAAJ">Junqi Gao</a><sup>4</sup>,
</span>
<span class="author-block">
<a href="https://scholar.google.com.hk/citations?user=sSOt_McAAAAJ">Dong Li</a><sup>4</sup>,
</span>
<span class="author-block">
<a href="https://dmksjfl.github.io">Jiafei Lyu</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="https://github.com/qianzhouyi2">Zhouyi Qian</a><sup>4</sup>,
</span>
<span class="author-block">
<a href="https://biqing-qi.github.io">Biqing Qi</a><sup>2†</sup>,
</span>
<span class="author-block">
<a href="https://www.sigs.tsinghua.edu.cn/lx/main.htm">Xiu Li</a><sup>1†</sup>,
</span>
<span class="author-block">
<a href="http://c3i.ee.tsinghua.edu.cn">Bowen Zhou</a><sup>1,2†</sup>
</span>
</div>
<div class="is-size-5 publication-authors" style="margin-top: 0.3em;">
<div class="author-block"><sup>1</sup>Tsinghua University, </div>
<div class="author-block"><sup>2</sup>Shanghai AI Laboratory, </div>
<div class="author-block"><sup>3</sup>BUPT, </div>
<div class="author-block"><sup>4</sup>Harbin Institute of Technology</div>
</div>
<div class="is-size-6 has-text-centered">
<!-- <p>*</p> -->
<div>*Equal contribution, †Corresponding author</div>
</div>
<!-- <div class="is-size-5 publication-authors" style="margin-top: 1em; color: orange;">
<div>Ranked <strong style="color: darkorange;">#1</strong> on HuggingFace Daily Papers</div>
</div> -->
<!-- Links -->
<div class="column has-text-centered" style="margin-top: 1em;">
<div class="publication-links">
<span class="link-block">
<a href="https://arxiv.org/pdf/2504.00891.pdf" class="external-link button is-normal is-rounded is-dark" target="_blank">
<span class="icon">
<i class="fas fa-file-pdf"></i>
</span>
<span>Paper</span>
</a>
</span>
<span class="link-block">
<a href="https://arxiv.org/abs/2504.00891" class="external-link button is-normal is-rounded is-dark" target="_blank">
<span class="icon">
<i class="ai ai-arxiv"></i>
</span>
<span>arXiv</span>
</a>
</span>
<span class="link-block">
<a href="https://github.com/RyanLiu112/GenPRM" class="external-link button is-normal is-rounded is-dark" target="_blank">
<span class="icon">
<i class="fab fa-github"></i>
</span>
<span>Code</span>
</a>
</span>
<span class="link-block">
<a href="https://huggingface.co/collections/GenPRM/genprm-67ee4936234ba5dd16bb9943" class="external-link button is-normal is-rounded is-dark" target="_blank">
<span class="icon">
<p style="font-size:14px">🤗</p>
</span>
<span>HuggingFace</span>
</a>
</span>
<span class="link-block">
<a href="https://github.com/RyanLiu112/Awesome-Process-Reward-Models" class="external-link button is-normal is-rounded is-dark" target="_blank">
<span class="icon">
<i class="fab fa-github"></i>
</span>
<span>Awesome Process Reward Models</span>
</a>
</span>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="hero teaser">
<div class="container is-max-desktop">
<div class="hero-body has-text-centered">
<img src="./static/images/comparison.png" alt="" style="max-width: 100%; height: auto;" id="comparison">
<h2 class="subtitle has-text-centered">
<strong>Figure 1</strong>: Comparison between GenPRM (right) and previous classification-based PRMs (left).
</h2>
</div>
</div>
</section>
<section class="hero teaser">
<div class="container is-max-desktop">
<div class="hero-body has-text-centered">
<img src="./static/images/framework2.png" alt="" style="max-width: 100%; height: auto;" id="framework">
<h2 class="subtitle has-text-centered">
<strong>Figure 2</strong>: Overall framework of GenPRM.
</h2>
</div>
</div>
</section>
<div id="results" class="container" style="margin-bottom: 2vh; max-width: 50%;">
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Results</h2>
<div class="content has-text-justified">
<p>
</p>
</div>
<img src="./static/images/main_processbench.png" alt="" style="width: 100%; max-width: 1000px; margin-top: 20px; margin-bottom: 10px;" id="main_processbench">
<div class="content has-text-justified">
<p>
</p>
</div>
<img src="./static/images/main_bon.png" alt="" style="width: 100%; max-width: 1000px; margin-top: 20px; margin-bottom: 10px;" id="main_bon">
<div class="content has-text-justified">
<p>
</p>
</div>
<img src="./static/images/critic.png" alt="" style="width: 100%; max-width: 1000px; margin-top: 20px; margin-bottom: 10px;" id="critic">
</div>
</div>
</div>
<section class="section" id="BibTeX">
<div class="container is-max-desktop content">
<div class="columns is-centered has-text-centered">
<h2 class="title">BibTeX</h2>
</div>
<pre><code>@article{zhao2025genprm,
title = {GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning},
author = {Jian Zhao and Runze Liu and Kaiyan Zhang and Zhimu Zhou and Junqi Gao and Dong Li and Jiafei Lyu and Zhouyi Qian and Biqing Qi and Xiu Li and Bowen Zhou},
journal = {arXiv preprint arXiv:2504.00891},
year = {2025}
}</code></pre>
Our collection of PRMs in <a href="https://github.com/RyanLiu112/Awesome-Process-Reward-Models">Awesome-Process-Reward-Models</a>:
<pre><code>@misc{Awesome-Process-Reward-Models,
title = {Awesome Process Reward Models},
author = {Runze Liu and Jian Zhao and Kaiyan Zhang and Zhimu Zhou and Junqi Gao and Dong Li and Jiafei Lyu and Zhouyi Qian and Biqing Qi and Xiu Li and Bowen Zhou},
howpublished = {\url{https://github.com/RyanLiu112/Awesome-Process-Reward-Models}},
note = {GitHub repository},
year = {2025}
}</code></pre>
Our recent work on LLM test-time scaling with PRMs:
<pre><code>@article{liu2025can,
title = {Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling},
author = {Runze Liu and Junqi Gao and Jian Zhao and Kaiyan Zhang and Xiu Li and Biqing Qi and Wanli Ouyang and Bowen Zhou},
journal = {arXiv preprint arXiv:2502.06703},
year = {2025}
}</code></pre>
</div>
</section>
<footer class="footer">
<div class="container">
<div class="columns is-centered">
<div class="column is-8">
<div class="content">
<p>
This website is licensed under a <a rel="license" href="http://creativecommons.org/licenses/by-sa/4.0/">Creative
Commons Attribution-ShareAlike 4.0 International License</a>.
</p>
<p>
This website template is adapted from <a rel="template"
href="https://nerfies.github.io/">Nerfies</a>.
</p>
</div>
</div>
</div>
</div>
</footer>
<style>
.shaded {
/* 10% opacity cyan */
background-color: rgba(0, 255, 255, 0.1);
}
</style>
</body>
</html>