-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
305 lines (272 loc) · 14.9 KB
/
index.html
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
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="description"
content="Deformable Neural Radiance Fields creates free-viewpoint portraits (nerfies) from casually captured videos.">
<meta name="keywords" content="Nerfies, D-NeRF, NeRF">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>READ</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.png">
<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>
<link href='https://fonts.googleapis.com/css?family=Chau Philomene One' rel='stylesheet'>
<link href='https://fonts.googleapis.com/css?family=EB Garamond' rel='stylesheet'>
<link href='https://fonts.googleapis.com/css?family=Montserrat' rel='stylesheet'>
<link href='https://fonts.googleapis.com/css?family=Anton' rel='stylesheet'>
<link href='https://fonts.googleapis.com/css?family=League Spartan' rel='stylesheet'>
<link href='https://fonts.googleapis.com/css?family=Bebas Neue' rel='stylesheet'>
<body>
<section class="hero">
<div class="hero-body" style="background-image: url(static/images/demo_thumbnail.png); background-size: cover; background-position: center; background-repeat: no-repeat; width: 100vw;">
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column has-text-centered">
<h1 class="title is-1 publication-title" style="font-family: League Spartan; color: white; font-size: 40px;">READ: Recurrent Adapter with Partial Video-Language Alignment for Parameter-Efficient Transfer Learning in Low-Resource Video-Language Modeling</h1>
<div class="is-size-5 publication-authors">
<span class="author-block" style="color: white !important; font-family: League Spartan; font-weight: bold;">
<a href="https://nguyentthong.github.io/" style="color: white !important; font-family: League Spartan; font-weight: bold; font-size: larger;">Thong Nguyen</a><sup style="font-family: League Spartan; font-weight: bold; color: white;">1</sup>,</span>
<span class="author-block" style="color: white !important; font-family: League Spartan; font-weight: bold;">
<a href="" style="color: white !important; font-family: League Spartan; font-weight: bold; font-size: larger">Xiaobao Wu</a><sup style="font-family: League Spartan; font-weight: bold; color: white">2</sup>,
</span>
<span class="author-block" style="color: white !important; font-family: League Spartan; font-weight: bold;">
<a href="" style="color: white !important; font-family: League Spartan; font-weight: bold; font-size: larger">Xinshuai Dong</a><sup style="font-family: League Spartan; font-weight: bold; color: white">3</sup>,
</span>
<span class="author-block" style="color: white !important; font-family: League Spartan; font-weight: bold;">
<a href="" style="color: white !important; font-family: League Spartan; font-weight: bold; font-size: larger">Khoi Le</a><sup style="font-family: League Spartan; font-weight: bold; color: white">4</sup>,
</span>
<span class="author-block" style="color: white !important; font-family: League Spartan; font-weight: bold;">
<a href="" style="color: white !important; font-family: League Spartan; font-weight: bold; font-size: larger">Zhiyuan Hu</a><sup style="font-family: League Spartan; font-weight: bold; color: white;">1</sup>,</span>
<span class="author-block" style="color: white !important; font-family: League Spartan; font-weight: bold;">
<a href="" style="color: white !important; font-family: League Spartan; font-weight: bold; font-size: larger">Cong-Duy Nguyen</a><sup style="font-family: League Spartan; font-weight: bold; color: white">2</sup>,
</span>
<span class="author-block" style="color: white !important; font-family: League Spartan; font-weight: bold;">
<a href="" style="color: white !important; font-family: League Spartan; font-weight: bold; font-size: larger">See-Kiong Ng</a><sup style="font-family: League Spartan; font-weight: bold; color: white">1</sup>,
</span>
<span class="author-block" style="color: white !important; font-family: League Spartan; font-weight: bold;">
<a href="" style="color: white !important; font-family: League Spartan; font-weight: bold; font-size: larger">Luu Anh Tuan</a><sup style="font-family: League Spartan; font-weight: bold; color: white">2</sup>,
</span>
</div>
<div class="is-size-5 publication-authors">
<span class="author-block" style="color: white; font-family: League Spartan; font-weight: bold; font-size: larger"><sup>1</sup>National University of Singapore (NUS),</span>
<span class="author-block" style="color: white; font-family: League Spartan; font-weight: bold; font-size: larger"><sup>2</sup>Nanyang Technological University (NTU),</span>
<span class="author-block" style="color: white; font-family: League Spartan; font-weight: bold; font-size: larger"><sup>3</sup>Carnegie Mellon University (CMU),</span>
<span class="author-block" style="color: white; font-family: League Spartan; font-weight: bold; font-size: larger"><sup>4</sup>VinAI Research</span>
</div>
<div class="column has-text-centered">
<div class="publication-links">
<span class="link-block">
<a href="https://arxiv.org/pdf/2312.06950"
class="external-link button is-normal is-rounded is-dark">
<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/2312.06950"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="ai ai-arxiv"></i>
</span>
<span>arXiv</span>
</a>
</span>
<span class="link-block">
<a href="https://github.com/nguyentthong/READ"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fab fa-github"></i>
</span>
<span>Code</span>
</a>
</span>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<br><br>
<section class="hero teaser">
<div class="container is-max-desktop">
<div class="hero-body">
<div class="container">
<h2 class="subtitle">
TLDR: Introduction of recurrent low-rank adaptation layers supported by an optimal-transport-based alignment objective between video and language representations in limited training data settings.
</h2>
</div>
</div>
</div>
</section>
<section class="section hero is-light">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3" style="color: #45322e; font-family: League Spartan;">Abstract</h2>
<div class="content has-text-justified">
<p>
Fully fine-tuning pretrained large-scale transformer models has become a popular paradigm for video-language modeling tasks, such as temporal language grounding and video-language summarization. With a growing number of tasks and limited training data, such full fine-tuning approach leads to costly model storage and unstable training. To overcome these shortcomings, we introduce lightweight adapters to the pre-trained model and only update them at fine-tuning time. However, existing adapters fail to capture intrinsic temporal relations among video frames or textual words. Moreover, they neglect the preservation of critical task-related information that flows from the raw video-language input into the adapter's low-dimensional space. To address these issues, we first propose a novel REcurrent ADapter (READ) that employs recurrent computation to enable temporal modeling capability. Second, we propose Partial Video-Language Alignment objective via the use of partial optimal transport to maintain task-related information flowing into our READ modules. We validate our READ framework through extensive experiments where READ significantly outperforms all existing fine-tuning strategies on multiple low-resource temporal language grounding and video-language summarization benchmarks.
</p>
</div>
</div>
</div>
</div>
</section>
<section class="section hero">
<div class="container is-max-desktop">
<!-- Title Section -->
<div class="columns is-centered has-text-centered">
<div class="column">
<h2 class="title is-3" style="font-family: League Spartan; color: #45322e;">Introduction</h2>
</div>
</div>
</div>
<br>
<div class="columns is-centered">
<div class="column is-three-fifths">
<ul style="text-align: left; list-style-type: disc;">
<li>Previous works fine-tune adaptation layers (Adapter) to mitigate the high computational cost and reduce overfitting as obtaining video-language data is expensive.</li>
<li>However, the adapters pay little attention to temporal modeling for video.</li>
<li>Moreover, the relationship between video and language modalities is also neglected. Such relationship is necessary to ground the entities or understand the language context.</li>
</ul>
</div>
</div>
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-full">
<img src="static/images/video_language_alignment_motivation.png" alt="JPG">
</div>
</div>
</div>
</div>
</section>
<hr class="solid">
<section class="section">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-full">
<h2 class="title is-3" style="font-family: League Spartan; color: #45322e;">Methodology</h2>
</div>
</div>
</div>
<div class="columns is-centered">
<div class="column is-three-fifths">
<ul style="text-align: left; list-style-type: disc;">
<li>We propose REcurrent ADapter (READ) to help low-rank adapters capture temporal information.</li>
<li>We integrate Partial Video-Language Alignment (PVLA) to encourage the alignment between video and language modalities.</li>
</ul>
</div>
</div>
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-full">
<img src="static/images/overall_architecture.png" alt="JPG">
</div>
</div>
</div>
</section>
<hr class="solid">
<section class="section">
<!-- 标题部分 -->
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-full">
<h2 class="title is-3" style="font-family: League Spartan; color: #45322e;">Experimental Results</h2>
</div>
</div>
</div>
<div class="columns is-centered">
<div class="column is-three-fifths">
<ul style="text-align: left; list-style-type: disc;">
<li>We evaluate our method on two prominent temporal grounding and video-text summarization tasks.</li>
</ul>
</div>
</div>
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-full">
<img src="static/images/experimental_results.png" alt="JPG">
</div>
</div>
</div>
<div class="columns is-centered">
<div class="column is-three-fifths">
<ul style="text-align: left; list-style-type: disc;">
<li>Our framework outperforms other fine-tuning approaches with significantly low trainable parameters.</li>
</ul>
</div>
</div>
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column is-full" style="width: 600px;">
<img src="static/images/performance_figure.png" alt="JPG">
</div>
</div>
</div>
</section>
<hr class="solid">
<!--BibTex citation -->
<section class="section" id="BibTeX">
<div class="container is-max-desktop content">
<h2 class="title has-text-centered" style="font-family: League Spartan; color: #45322e;">BibTeX</h2>
<pre><code>
@article{nguyen2024read,
author = {Nguyen, Thong and Wu, Xiaobao and Dong, Xinshuai and Le, Khoi M and Hu, Zhiyuan and Nguyen, Cong-Duy and Ng, See-Kiong and Luu, Anh Tuan},
title = {READ: Recurrent Adapter with Partial Video-Language Alignment for Parameter-Efficient Transfer Learning in Low-Resource Video-Language Modeling},
journal = {AAAI},
year = {2024},
}
</code></pre>
</div>
</section>
<footer class="footer">
<div class="container">
<div class="content has-text-centered">
<a class="icon-link"
href="https://arxiv.org/pdf/2312.06950">
<i class="fas fa-file-pdf"></i>
</a>
<a class="icon-link" href="https://github.com/nguyentthong/READ" class="external-link" disabled>
<i class="fab fa-github"></i>
</a>
</div>
<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>
</div>
</div>
</div>
</div>
</footer>
</body>
</html>