-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
327 lines (323 loc) · 11.2 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
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
<!DOCTYPE html>
<html>
<head>
<link rel="stylesheet" type="text/css" href="css/bootstrap.css">
<link rel="stylesheet" type="text/css" href="css/style.css">
<link href="https://fonts.googleapis.com/css?family=Raleway:100,200,300,400,500,600,700,800,900" rel="stylesheet">
<link rel="stylesheet" href="css/fontAwesome.css">
<title> D4 - a Chinese Dialogue Dataset for Depression-Diagnosis-Oriented Chat </title>
</head>
<body>
<header class="header">
<div class="container clearfix">
<div class="left-container">
<a href="#"> D<sup>4</sup> </a>
</div>
<div class="right-container">
<ul class="nav-menu clearfix">
<li>
<a href="#"> Home </a>
<li>
<!-- <li>
<a href="dashboard.html"> Explore v1.0 </a>
</li> -->
</ul>
</div>
</div>
</header>
<section class="banner">
<div class="container clearfix">
<h1> D<sup>4</sup> </h1>
<span> A Chinese Dialogue Dataset for Depression-Diagnosis-Oriented Chat </span>
<!-- <a href="data.html" class="download-hyperlink"><button class="download-button">
<i class="fa fa-arrow-circle-o-down" aria-hidden="true"></i> Download
</button>
</a> -->
</div>
</section>
<section class="main-content">
<div class="container clearfix">
<div class="row">
<div class="col-md-5">
<div class="info-container">
<span class="info-title">
<i class="fa fa-info-circle" aria-hidden="true"></i>
What is D<sup>4</sup>?
</span>
<p>
D<sup>4</sup> is a novel Chinese <strong>D</strong>ialogue <strong>D</strong>ataset for <strong>D</strong>epression-<strong>D</strong>iagnosis-Oriented Chat, which is a brand new type of dialogue called Task-Oriented Chat. It consists of 1,339 multi-turn (Avg.turns 21.6) dialogues with dialogue summary and diagnosis results. This dataset mainly serves for constructing a more
empathy-driven and diagnostic-accurate consultation dialogue system.
</p>
<button class="action-button" style="display: none;">
Button 1
</button>
<button class="inverse-button" style="display: none;">
Button 2
</button>
<span class="info-title">
<i class="fa fa-exclamation-triangle" aria-hidden="true"></i>
Get Access
</span>
<p>
To get access to the data, you must <strong>Sign a Data Usage Agreement (DUA) </strong> to prevent the invasion of privacy or other potential misuses.
<span>
<br/>Please read the DUA carefully, and especially <strong>Pay Attention</strong> that:
<!-- <br/> -->
<br/>1. You cannot transfer or share any part of the dataset.
<br/>2. You cannot identify or contact any user in the dataset.
<br/>3. You can only use the data for <strong>research purposes</strong>.
<!-- <br/> -->
<br/>Please send a email to <a href="mailto:[email protected]">[email protected]</a> with the message: "<i>I consent to the Data Usage Agreement (DUA)</i>." and attach the DUA including your <strong> handwritten signature </strong> in it.
<br/>
<br/>
</span>
<a href="https://docs.google.com/document/d/1dcT4Z4uK4f9nfoOhRvUR6uv-KH1SRdh6/edit?usp=sharing&ouid=117307988136809658128&rtpof=true&sd=true" class="download-link">
<i class="fa fa-google" aria-hidden="true"></i>
Download DUA from Google Drive
</a>
</p>
<span class="info-title">
<i class="fa fa-star-o" aria-hidden="true"></i>
News
</span>
<p class="row">
<!-- <span class="news-date col-md-2"> Oct 22, 2022 </span> -->
<!-- <span class="news-context"> The full dataset and baseline is available in: </span>
<a href="https://pan.baidu.com/s/1-sqWgc8bTm6-bkAFYJwYQQ" class="download-link">
<i class="fa fa-cloud-download" aria-hidden="true"></i>
Extraction code: <strong>2mys</strong>
</a>
<a href="https://websrc-data.s3.amazonaws.com/release.zip" class="download-link">
<i class="fa fa-amazon" aria-hidden="true"></i>
Download from Amazon.com
</a> -->
<!-- <span class="news-date col-md-2"> Oct 27, 2022 </span>
<span class="news-context"> The baseline and code are available at
<a href="https://github.com/BigBinnie/D4_baseline" class="download-link">
<i class="fa fa-github" aria-hidden="true"></i>
D4_baseline
</a> </span> -->
<span class="news-date col-md-2"> Oct 6, 2022 </span>
<span class="news-context"> Our paper is accepted by EMNLP 2022. </span>
<span class="news-date col-md-2"> Oct 27, 2022 </span>
<span class="news-context"> Baseline is available on <br/>
<a href="https://github.com/BigBinnie/D4_baseline" class="download-link">
<i class="fa fa-github" aria-hidden="true"></i>
D4_baseline
</a></span>
</p>
<span class="info-title">
<i class="fa fa-question-circle-o" aria-hidden="true"></i>
Contact Us
</span>
<p>
If you have any questions about this dataset, please contact
<a href="mailto:[email protected]">[email protected]</a>,
<a href="mailto:[email protected]">[email protected]</a>
or <a href="mailto:[email protected]">[email protected]</a>
</p>
</div>
</div>
<div class="col-md-7">
<div class="info-container">
<span class="info-title">
<i class="fa fa-database" aria-hidden="true"></i>
Statistics
</span>
<p class="row">
<strong> Depression Severity Stats</strong>
<table class="leader-board">
<tbody>
<tr>
<th> Category </th>
<th> Overall</th>
<th> Control </th>
<th> Mild </th>
<th> Moderate </th>
<th> Severe </th>
</tr>
<tr>
<th> Dialogues </th>
<th>1,339</th>
<th> 430 </th>
<th> 342 </th>
<th> 368 </th>
<th> 199 </th>
</tr>
<tr>
<th> Avg. Turns </th>
<th> 21.6</th>
<th> 17.9 </th>
<th> 21.3 </th>
<th> 23.7 </th>
<th> 26.0 </th>
</tr>
<tr>
<th> Avg. Tokens of Symptom Summary </th>
<th> 84.4 </th>
<th> 59.8 </th>
<th> 82.0 </th>
<th> 100.5 </th>
<th> 111.9 </th>
</tr>
</tbody>
</table>
</p>
<br/>
<strong> Topic Stats</strong>
<p text-aligh="center">
<div class="row">
<div class="col-md-6"><img src="image/topic ratio.png " alt="Topic Ratio" width="100%"></div>
<div class="col-md-6"><img src="image/6topic.png " alt="Topic Ratio" width="100%"></div>
</div>
</p>
<span class="info-title">
<i class="fa fa-bar-chart" aria-hidden="true"></i>
Leaderboard
</span>
<p>
4 tasks on D4 introduced in our paper, including Response Generation, Topic Prediction, Dialog Summary and Severity Classification. The baseline will be available soon.
</p>
<!-- <table class="leader-board">
<tbody>
<tr>
<th> Rank </th>
<th> Model </th>
<th> EM </th>
<th> F1 </th>
<th> POS </th>
</tr>
<tr>
<td>
1
<span>
Oct 12, 2022
</span>
</td>
<td>
DocPrompt (ErnieLayout-Large)
<p> BAIDU-Document Intelligence </p>
<a class='link'> (Wu et al.) </a>
<a class='link' href='https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-layoutx'> code </a>
<a class='link' href='https://huggingface.co/spaces/PaddlePaddle/ERNIE-LayoutX'> demo </a>
</td>
<td class="stat"> 77.35 </td>
<td class="stat"> 85.04 </td>
<td class="stat"> N/A </td>
</tr>
<tr>
<td>
2
<span>
Mar 01, 2022
</span>
</td>
<td>
TIE (MarkupLM-Large)
<p> Shanghai Jiao Tong University </p>
<a class='link' href='https://arxiv.org/abs/2205.06435'> (Zhao et al., NAACL'22) </a>
<a class='link' href='https://github.com/X-LANCE/TIE'> code </a>
</td>
<td class="stat"> 76.25 </td>
<td class="stat"> 80.51 </td>
<td class="stat"> 89.50 </td>
</tr>
<tr>
<td>
3
<span>
Mar 01, 2022
</span>
</td>
<td>
TIE (MarkupLM-Large)-3_seeds_ave
<p> Shanghai Jiao Tong University </p>
<a class='link' href='https://arxiv.org/abs/2205.06435'> (Zhao et al., NAACL'22) </a>
<a class='link' href='https://github.com/X-LANCE/TIE'> code </a>
</td>
<td class="stat"> 75.87 </td>
<td class="stat"> 80.19 </td>
<td class="stat"> 89.73 </td>
</tr>
<tr>
<td>
4
<span>
Nov 30, 2021
</span>
</td>
<td>
MarkupLM-Large
<p> MSRA+SJTU </p>
<a class='link' href='https://arxiv.org/abs/2110.08518'> (Li et al., ACL'22) </a>
<a class='link' href='https://github.com/microsoft/unilm/tree/master/markuplm'> code </a>
</td>
<td class="stat"> 69.87 </td>
<td class="stat"> 77.94 </td>
<td class="stat"> 88.09 </td>
</tr>
<tr>
<td>
5
<span>
Nov 30, 2021
</span>
</td>
<td>
MarkupLM-Large (3_seeds_ave)
<p> MSRA+SJTU </p>
<a class='link' href='https://arxiv.org/abs/2110.08518'> (Li et al., ACL'22) </a>
<a class='link' href='https://github.com/microsoft/unilm/tree/master/markuplm'> code </a>
</td>
<td class="stat"> 69.09 </td>
<td class="stat"> 76.45 </td>
<td class="stat"> 87.24 </td>
</tr>
<tr>
<td>
6
<span>
Sept 01, 2021
</span>
</td>
<td>
V-PLM (ELECTRA)
<p> Shanghai Jiao Tong University </p>
<a class='link' href='https://arxiv.org/abs/2101.09465'> (Chen et al., EMNLP'21) </a>
<a class='link' href='https://github.com/X-LANCE/WebSRC-Baseline'> code </a>
</td>
<td class="stat"> 68.07 </td>
<td class="stat"> 75.25 </td>
<td class="stat"> 84.96 </td>
</tr>
<tr>
<td>
7
<span>
Sept 01, 2021
</span>
</td>
<td>
V-PLM (BERT)
<p> Shanghai Jiao Tong University </p>
<a class='link' href='https://arxiv.org/abs/2101.09465'> (Chen et al., EMNLP'21) </a>
<a class='link' href='https://github.com/X-LANCE/WebSRC-Baseline'> code </a>
</td>
<td class="stat"> 54.84 </td>
<td class="stat"> 62.80 </td>
<td class="stat"> 76.39 </td>
</tr>
</tbody>
</table> -->
</div>
</div>
</div>
</div>
</section>
<section class="footer clearfix">
<a href="http://x-lance.sjtu.edu.cn/"> SJTU X-LANCE Lab<a>
</section>
<script src="js/jquery-1.11.2.min.js"></script>
<script type="text/javascript" src="js/footer.js"></script>
</body>
</html>