-
Notifications
You must be signed in to change notification settings - Fork 2
/
index.html
294 lines (250 loc) · 14.6 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
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
<meta name="description" content="">
<meta name="author" content="">
<title>MedMCQA Homepage</title>
<!-- Bootstrap core CSS -->
<link href="vendor/bootstrap/css/bootstrap.min.css" rel="stylesheet">
<!-- Custom fonts for this template -->
<link href="vendor/font-awesome/css/font-awesome.min.css" rel="stylesheet" type="text/css">
<link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.3.1/css/all.css" integrity="sha384-mzrmE5qonljUremFsqc01SB46JvROS7bZs3IO2EmfFsd15uHvIt+Y8vEf7N7fWAU" crossorigin="anonymous">
<link href='https://fonts.googleapis.com/css?family=Lora:400,700,400italic,700italic' rel='stylesheet' type='text/css'>
<link href='https://fonts.googleapis.com/css?family=Open+Sans:300italic,400italic,600italic,700italic,800italic,400,300,600,700,800' rel='stylesheet' type='text/css'>
<!-- Custom styles for this template -->
<link href="css/clean-blog.min.css" rel="stylesheet">
<link href="css/MedMCQA.css" rel="stylesheet">
</head>
<body>
<nav class="navbar navbar-expand-lg navbar-light fixed-top" id="mainNav">
<div class="container">
<a class="navbar-brand" href="index.html">MedMCQA</a>
<button class="navbar-toggler navbar-toggler-right" type="button" data-toggle="collapse" data-target="#navbarResponsive" aria-controls="navbarResponsive" aria-expanded="false" aria-label="Toggle navigation">
Menu
<i class="fa fa-bars"></i>
</button>
</div>
</nav>
<!-- Page Header -->
<header class="masthead" style="background-image: url('img/background.png')">
<div class="overlay"></div>
<div class="container">
<div class="row">
<div class="col-lg-8 col-md-10 mx-auto">
<div class="site-heading">
<h1>MedMCQA</h1>
<span class="subheading">Multi-Subject Multi-Choice Dataset
for Medical domain</span>
</div>
</div>
</div>
</div>
</header>
<!-- Main Content -->
<div class="container" id="main">
<div class="row">
<div class="col-lg-5">
<div class="list-group">
<div class="list-group-item"><h4>About</h4>
<p></p>
<p> MedMCQA, a large-scale, Multiple-Choice Question Answering (MCQA) dataset designed to address realworld medical entrance exam questions. <br> The MedMCQA task can be formulated as X = {Q, O} where Q represents the questions in the text, O represents the candidate options, multiple candidate answers are given for each question O = {O1, O2, ..., On}. The goal is to select the single or multiple answers from the option set.</p>
<p class="text-center">
<a href="https://proceedings.mlr.press/v174/pal22a.html"><button type="button" class="btn btn-info">MedMCQA Paper (ACM CHIL'22)</button></a>
</p>
</div>
<div class="list-group-item"><h4>Dataset</h4>
<p>MedMCQA has More than 194k high-quality AIIMS & NEET PG entrance exam MCQs covering 2.4k healthcare topics and 21 medical subjects are collected
with an average token length of 12.77 and high topical diversity. </p>
<p class="text-center">
<a href="https://github.com/MedMCQA/MedMCQA"><button type="button" class="btn btn-success">MedMCQA Data & Code Repository</button></a>
</p>
</div>
<div class="list-group-item"><h4>Submission</h4>
<p>To submit your model, please follow the instructions in the GitHub repository.</p>
</div>
<div class="list-group-item"><h4>Citation</h4>
<p>If you use MedMCQA in your research, please cite our paper by:</p>
<pre>
@InProceedings{pmlr-v174-pal22a,
title = {MedMCQA: A Large-scale Multi-Subject Multi-Choice Dataset for Medical domain Question Answering},
author = {Pal, Ankit and Umapathi, Logesh Kumar and Sankarasubbu, Malaikannan},
booktitle = {Proceedings of the Conference on Health, Inference, and Learning},
pages = {248--260},
year = {2022},
editor = {Flores, Gerardo and Chen, George H and Pollard, Tom and Ho, Joyce C and Naumann, Tristan},
volume = {174},
series = {Proceedings of Machine Learning Research},
month = {07--08 Apr},
publisher = {PMLR},
pdf = {https://proceedings.mlr.press/v174/pal22a/pal22a.pdf},
url = {https://proceedings.mlr.press/v174/pal22a.html},
abstract = {This paper introduces MedMCQA, a new large-scale, Multiple-Choice Question Answering (MCQA) dataset designed to address real-world medical entrance exam questions. More than 194k high-quality AIIMS & NEET PG entrance exam MCQs covering 2.4k healthcare topics and 21 medical subjects are collected with an average token length of 12.77 and high topical diversity. Each sample contains a question, correct answer(s), and other options which requires a deeper language understanding as it tests the 10+ reasoning abilities of a model across a wide range of medical subjects & topics. A detailed explanation of the solution, along with the above information, is provided in this study.}
}
</pre>
</div>
</div>
</div>
<div class="col-lg-7">
<div class="card card-outline-secondary">
<div class="card-header">
Leaderboard (w/o Context)
</div>
<div class="card-body">
In the <em>w/o Context</em> setting, For the experiments that do not use context, <br>
<span class="badge badge-secondary">[CLS]</span> Question <span class="badge badge-secondary">[SEP]</span> Option <span class="badge badge-secondary">[SEP]</span>
<table class="table table-responsive">
<thead class="thead-light">
<tr>
<th scope="col" rowspan=2></th>
<th scope="col" rowspan=2 class="align-middle text-center">Model</th>
<th scope="col" rowspan=2 class="align-middle text-center">Code</th>
<th scope="col" colspan=1 class='text-center'>Test Set</th>
<th scope="col" colspan=1 class='text-center'>Dev Set</th>
</tr>
<tr>
<th scope="col" class='text-center'>Acc (%)</th>
<th scope="col" class='text-center'>Acc (%)</th>
</tr>
</thead>
<tbody><tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">March 10, 2022</span></td>
<td class='align-middle text-center'> BERT (Devlin et al., 2019) <sub>Base</sub><br/></td>
<td class='align-middle text-center'><a href="https://github.com/medmcqa/medmcqa"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.33</strong></td>
<td class='align-middle text-center'><strong>0.35</strong></td>
</tr>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">March 10, 2022</span></td>
<td class='align-middle text-center'> BioBERT (Lee et al.,2020) <br/></td>
<td class='align-middle text-center'><a href="https://github.com/medmcqa/medmcqa"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.37</strong></td>
<td class='align-middle text-center'><strong>0.38</strong></td>
</tr>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">March 10, 2022</span></td>
<td class='align-middle text-center'> SciBERT (Beltagy et al., 2019) <br/></td>
<td class='align-middle text-center'><a href="https://github.com/medmcqa/medmcqa"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.39</strong></td>
<td class='align-middle text-center'><strong>0.39</strong></td>
</tr>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">March 10, 2022</span></td>
<td class='align-middle text-center'> PubmedBERT(Gu et al., 2022) <br/></td>
<td class='align-middle text-center'><a href="https://github.com/medmcqa/medmcqa"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.41</strong></td>
<td class='align-middle text-center'><strong>0.40</strong></td>
</tr>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">December 5, 2022</span></td>
<td class='align-middle text-center'> Codex <small>5-shot CoT</small> (<a href="https://arxiv.org/abs/2207.08143">Liévin et al., 2022</a>) <br/></td>
<td class='align-middle text-center'><a href="https://github.com/vlievin/medical-reasoning"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.60</strong></td>
<td class='align-middle text-center'><strong>0.63</strong></td>
</tr>
</tbody>
</table>
</div>
</div>
<div class="card card-outline-secondary">
<div class="card-header">
Leaderboard (with Context)
</div>
<div class="card-body">
In the <em>with Context</em> setting, These contexts are combined by [SEP] token with the concatenation of question and answer pair. This creates four input sequences per question. <br>
<span class="badge badge-secondary">[CLS]</span> Context <span class="badge badge-secondary">[SEP]</span> Question <span class="badge badge-secondary">[SEP]</span> Option <span class="badge badge-secondary">[SEP]</span> <br>
<table class="table table-responsive">
<thead class="thead-light">
<tr>
<th scope="col" rowspan=2></th>
<th scope="col" rowspan=2 class="align-middle text-center">Model</th>
<th scope="col" rowspan=2 class="align-middle text-center">Code</th>
<th scope="col" colspan=1 class='text-center'>Test Set</th>
<th scope="col" colspan=1 class='text-center'>Dev Set</th>
</tr>
<tr>
<th scope="col" class='text-center'>Acc (%)</th>
<th scope="col" class='text-center'>Acc (%)</th>
</tr>
</thead>
<tbody>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">March 10, 2022</span></td>
<td class='align-middle text-center'> BERT (Devlin et al., 2019) <sub>Base</sub><br/></td>
<td class='align-middle text-center'><a href="https://github.com/medmcqa/medmcqa"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.37</strong></td>
<td class='align-middle text-center'><strong>0.35</strong></td>
</tr>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">March 10, 2022</span></td>
<td class='align-middle text-center'> BioBERT (Lee et al.,2020)<br/></td>
<td class='align-middle text-center'><a href="https://github.com/medmcqa/medmcqa"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.42</strong></td>
<td class='align-middle text-center'><strong>0.39</strong></td>
</tr>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">March 10, 2022</span></td>
<td class='align-middle text-center'> SciBERT (Beltagy et al., 2019) <br/></td>
<td class='align-middle text-center'><a href="https://github.com/medmcqa/medmcqa"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.43</strong></td>
<td class='align-middle text-center'><strong>0.41</strong></td>
</tr>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">March 10, 2022</span></td>
<td class='align-middle text-center'> PubmedBERT(Gu et al., 2022) <br/></td>
<td class='align-middle text-center'><a href="https://github.com/medmcqa/medmcqa"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.47</strong></td>
<td class='align-middle text-center'><strong>0.43</strong></td>
</tr>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">July 17, 2022</span></td>
<td class='align-middle text-center'> InstructGPT <small>zero-shot CoT</small> (<a href="https://arxiv.org/abs/2207.08143">Liévin et al., 2022</a>) <br/></td>
<td class='align-middle text-center'><a href="https://github.com/vlievin/medical-reasoning"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.49</strong></td>
<td class='align-middle text-center'><strong>0.49</strong></td>
</tr>
<tr>
<td scope="row" class='align-middle text-center'>1<br>
<span class="badge badge-secondary">September 23, 2022</span></td>
<td class='align-middle text-center'> VOD <small>BioLinkBERT</small> (<a href="https://arxiv.org/abs/2210.06345">Liévin et al., 2022</a>) <br/></td>
<td class='align-middle text-center'><a href="https://github.com/VodLM/vod-lm"><i class="far fa-file-alt"></i></a></td>
<td class='align-middle text-center'><strong>0.58</strong></td>
<td class='align-middle text-center'><strong>0.63</strong></td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
</div>
</div>
<hr>
<!-- Footer -->
<footer>
<div class="container">
<div class="row">
<div class="col-lg-8 col-md-10 mx-auto">
<p class="copyright text-muted">Leaderboard template kindly provided by the HotpotQA team.</p>
<p class="copyright text-muted">Theme adapted from Start Bootstrap's <a href="https://startbootstrap.com/template-overviews/clean-blog/">Clean Blog</a> template.</p>
</div>
</div>
</div>
</footer>
<!-- Bootstrap core JavaScript -->
<script src="vendor/jquery/jquery.min.js"></script>
<script src="vendor/bootstrap/js/bootstrap.bundle.min.js"></script>
<!-- Custom scripts for this template -->
<script src="js/clean-blog.min.js"></script>
</body>
</html>