-
Notifications
You must be signed in to change notification settings - Fork 2
/
index.html
448 lines (394 loc) · 20.6 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
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
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<!-- Here are the current paper tags:
1) tag-all (for all papers)
2) tag-privacy
3) tag-host-security
4) tag-network-security
5) tag-theoretical-foundations
-->
<html><head>
<title>Deep Learning for Program Synthesis</title>
<style type="text/css">
body {
margin-top: 30px;
margin-bottom: 30px;
margin-left: 100px;
margin-right: 100px;
}
p {
margin-top: 0px;
margin-bottom: 0px;
}
.caption {
font-size: 34px;
font-weight: normal;
color: #000;
font-family: Constantia, "Lucida Bright", "DejaVu Serif", Georgia, serif;
}
.caption-1 {
font-size: 16px;
font-family: Tahoma, Geneva, sans-serif;
}
.caption-2 {
font-size: 16px;
font-family: Tahoma, Geneva, sans-serif;
font-weight: bold;
color: #990000;
}
.caption-3 {
font-size: 16px;
font-family: Tahoma, Geneva, sans-serif;
font-weight: bold;
color: #F00;
}
.caption-4 {
font-size: 16px;
font-family: Tahoma, Geneva, sans-serif;
color: #990000;
}
.content {
font-size: 16px;
font-family: Tahoma, Geneva, sans-serif;
text-align: justify;
}
.title-small {
font-size: 20px;
font-family: Georgia, "Times New Roman", Times, serif;
font-weight: bold;
color: #F90;
}
.title-large {
font-size: 28px;
font-family: Georgia, "Times New Roman", Times, serif;
font-weight: bold;
color: #000;
}
.margin {
font-size: 10px;
line-height: 10px;
}
.margin-small {
font-size: 5px;
line-height: 5px;
}
.margin-large {
font-size: 16px;
line-height: 16px;
}
</style>
<script type="text/javascript" src="jquery.js"></script>
<script type="text/javascript">
function displaypage(){
$(".tag-all").hide();
if(document.getElementById('tag-all-box').checked){
$(".tag-all").show();
}
if(document.getElementById('tag-theoretical-foundations-box').checked){
$(".tag-theoretical-foundations").show();
}
if(document.getElementById('tag-host-security-box').checked){
$(".tag-host-security").show();
}
if(document.getElementById('tag-network-security-box').checked){
$(".tag-network-security").show();
}
if(document.getElementById('tag-privacy-box').checked){
$(".tag-privacy").show();
}
}
$(document).ready(function() {
$("#tag-all-box").click(function() {
displaypage();
});
$("#tag-theoretical-foundations-box").click(function() {
displaypage();
});
$("#tag-host-security-box").click(function() {
displaypage();
});
$("#tag-network-security-box").click(function() {
displaypage();
});
$("#tag-privacy-box").click(function() {
displaypage();
});
});
</script>
<meta content="text/html; charset=unicode" http-equiv="Content-Type">
<meta name="GENERATOR" content="MSHTML 9.00.8112.16443"></head>
<body><center><h1><span>Deep Learning for Program Synthesis</span></h1></center>
<center>
[<a href="#statement">Research Statement</a>] [<a href ="#publications">Publications</a>] [<a href="#members">Members</a>]
</center>
<h2 class="label"><a name="statement"><span >Research Statement</span></a></h2>
Synthesizing a program from a specification has been a long-standing challenge. Recent research have demonstrated that deep neural networks have the potential to learn a program satisfying various specification methods, such as natural language descriptions and input-output examples. The ability to automatically synthesize code has numerous applications, ranging from helping end-users (non-technical users) write programs, helping software developers synthesize mundane pieces of code or optimized code, helping data scientists clean up and explore data, to helping algorithm designers discover new algorithms.
<br><br>
This problem is extremely challenging, and the complexity of the synthesized programs by existing approaches is still limited. In our research for neural program synthesis, we aim at generating programs with more complexity, better generalizability, while guaranteeing the correctness. In this process, besides enabling real-world applications using program synthesis, we hope to make contributions towards addressing core challenges in deep learning including generalization, search, abstraction, and representation. We believe that solving the program synthesis problem is a good step towards solving AGI (artificial general intelligence).
<hr>
<h2 class="label"><a name="publications"><span >Recent Publications</span></a></h2>
<!--<form>
<div id="tags">
<input type="radio" name="tags" checked id="tag-all-box">
<label for="tag-all-box">All Publications</label>
<input type="radio" name="tags" id="tag-theoretical-foundations-box">
<label for="tag-theoretical-foundations-box">Theoretical Foundations</label>
<input type="radio" name="tags" id="tag-host-security-box">
<label for="tag-host-security-box">Host Security</label>
<input type="radio" name="tags" id="tag-network-security-box">
<label for="tag-network-security-box">Network Security</label>
<input type="radio" name="tags" id="tag-privacy-box">
<label for="tag-privacy-box">Privacy</label>
</div>
</form> -->
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/codesyntax.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/2210.14473"><strong>Benchmarking Language Models for Code Syntax Understanding</strong></a></p>
<p class="content">Da Shen, Xinyun Chen, Chenguang Wang, Koushik Sen, Dawn Song.</p>
<p class="content">Findings of Conference on Empirical Methods in Natural Language Processing (EMNLP Findings). December, 2022.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/latent-execution.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/2107.00101"><strong>Latent Execution for Neural Program Synthesis Beyond Domain-Specific Languages</strong></a></p>
<p class="content">Xinyun Chen, Dawn Song, Yuandong Tian.</p>
<p class="content">Advances in Neural Information Processing Systems (NeurIPS). December, 2021.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/plotcoder.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://aclanthology.org/2021.acl-long.169/"><strong>PlotCoder: Hierarchical Decoding for Synthesizing Visualization Code in Programmatic Context</strong></a></p>
<p class="content">Xinyun Chen, Linyuan Gong, Alvin Cheung, Dawn Song.</p>
<p class="content">Annual Meeting of the Association for Computational Linguistics (ACL). August, 2021.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/spidersyn.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://aclanthology.org/2021.acl-long.195/"><strong>Towards Robustness of Text-to-SQL Models against Synonym Substitution</strong></a></p>
<p class="content">Yujian Gan, Xinyun Chen, Qiuping Huang, Matthew Purver, John R. Woodward, Jinxia Xie, Pengsheng Huang.</p>
<p class="content">Annual Meeting of the Association for Computational Linguistics (ACL). August, 2021.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/spreadsheetcoder.gif" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="http://proceedings.mlr.press/v139/ren21a.html"><strong>SpreadsheetCoder: Formula Prediction from Semi-structured Context</strong></a></p>
<p class="content">Xinyun Chen, Petros Maniatis, Rishabh Singh, Charles Sutton, Hanjun Dai, Max Lin, Denny Zhou.</p>
<p class="content">International Conference on Machine Learning (ICML). July, 2021.</p>
<p class="content">
<a href="https://workspaceupdates.googleblog.com/2021/08/intelligent-formula-and-function-suggestions-in-google-sheets.html">Google Workspace Update</a>
</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/lego.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="http://proceedings.mlr.press/v139/ren21a.html"><strong>LEGO: Latent Execution-Guided Reasoning for Multi-Hop Question Answering on Knowledge Graphs</strong></a></p>
<p class="content">Hongyu Ren, Hanjun Dai, Bo Dai, Xinyun Chen, Michihiro Yasunaga, Haitian Sun, Dale Schuurmans, Jure Leskovec, Denny Zhou.</p>
<p class="content">International Conference on Machine Learning (ICML). July, 2021.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/apps.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/2105.09938"><strong>Measuring Coding Challenge Competence With APPS</strong></a></p>
<p class="content">Dan Hendrycks*, Steven Basart*, Saurav Kadavath, Mantas Mazeika, Akul Arora, Ethan Guo, Collin Burns, Samir Puranik, Horace He, Dawn Song, Jacob Steinhardt.</p>
<p class="content">May, 2021.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/ness.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/2008.06662"><strong>Compositional Generalization via Neural-Symbolic Stack Machines</strong></a></p>
<p class="content">Xinyun Chen, Chen Liang, Adams Wei Yu, Dawn Song, Denny Zhou.</p>
<p class="content">Advances in Neural Information Processing Systems (NeurIPS). December, 2020.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/sed.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/2007.08095"><strong>Synthesize, Execute and Debug: Learning to Repair for Neural Program Synthesis</strong></a></p>
<p class="content">Kavi Gupta, Peter Ebert Christensen*, Xinyun Chen*, Dawn Song.</p>
<p class="content">Advances in Neural Information Processing Systems (NeurIPS). December, 2020.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/201911-ratsql.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/1911.04942"><strong>RAT-SQL: Relation-Aware Schema Encoding and Linking for Text-to-SQL Parsers</strong></a></p>
<p class="content">Bailin Wang*, Richard Shin*, Xiaodong Liu, Oleksandr Polozov, Matthew Richardson.</p>
<p class="content">Annual Meeting of the Association for Computational Linguistics (ACL). July, 2020.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/202006-nerd.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://openreview.net/forum?id=ryxjnREFwH"><strong>Neural Symbolic Reader: Scalable Integration of Distributed and Symbolic Representations for Reading Comprehension</strong></a></p>
<p class="content">Xinyun Chen, Chen Liang, Adams Wei Yu, Denny Zhou, Dawn Song, Quoc V. Le.</p>
<p class="content">International Conference on Learning Representations (ICLR). May, 2020.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/202006-superopt.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://openreview.net/forum?id=r1egIyBFPS"><strong>Deep Symbolic Superoptimization Without Human Knowledge</strong></a></p>
<p class="content">Hui Shi, Yang Zhang, Xinyun Chen, Yuandong Tian, Jishen Zhao.</p>
<p class="content">International Conference on Learning Representations (ICLR). May, 2020.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/201912-coda.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/1906.12029"><strong>Coda: An End-to-End Neural Program Decompiler</strong></a></p>
<p class="content">Cheng Fu, Huili Chen, Haolan Liu, Xinyun Chen, Yuandong Tian, Farinaz Koushanfar, Jishen Zhao.</p>
<p class="content">Advances in Neural Information Processing Systems (NeurIPS). December, 2019.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/201912-idioms.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/1906.10816"><strong>Program Synthesis and Semantic Parsing with Learned Code Idioms</strong></a></p>
<p class="content">Richard Shin, Marc Brockschmidt, Militadis Allamanis, Oleksandr Polozov.</p>
<p class="content">Advances in Neural Information Processing Systems (NeurIPS). December, 2019.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/exec.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://openreview.net/forum?id=H1gfOiAqYm"><strong>Execution-Guided Neural Program Synthesis</strong></a></p>
<p class="content">Xinyun Chen, Chang Liu, Dawn Song.</p>
<p class="content">International Conference on Learning Representations (ICLR). May, 2019.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/karel-analysis.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://openreview.net/forum?id=ryeOSnAqYm"><strong>Synthetic Datasets for Neural Program Synthesis</strong></a></p>
<p class="content">Richard Shin, Neel Kant, Kavi Gupta, Chris Bender, Brandon Trabucco, Rishabh Singh, Dawn Song.</p>
<p class="content">International Conference on Learning Representations (ICLR). May, 2019.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/karel-trace-to-code.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://papers.nips.cc/paper/8107-improving-neural-program-synthesis-with-inferred-execution-traces"><strong>Improving Neural Program Synthesis with Inferred Execution Traces</strong></a></p>
<p class="content">Richard Shin, Illia Polosukhin, Dawn Song.</p>
<p class="content">Advances in Neural Information Processing Systems (NIPS). December, 2018.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/tree2tree.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/1802.03691"><strong>Tree-to-tree Neural Networks for Program Translation</strong></a></p>
<p class="content">Xinyun Chen, Chang Liu, Dawn Song.</p>
<p class="content">Advances in Neural Information Processing Systems (NIPS). December, 2018.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/neuralParser.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/1706.01284"><strong>Towards Synthesizing Complex Programs from Input-Output Examples</strong></a></p>
<p class="content">Xinyun Chen, Chang Liu, Dawn Song.</p>
<p class="content">International Conference on Learning Representations (ICLR). May, 2018.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/php.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://openreview.net/forum?id=rJl63fZRb"><strong>Parametrized Hierarchical Procedures for Neural Programming</strong></a></p>
<p class="content">Roy Fox, Richard Shin, Sanjay Krishnan, Ken Goldberg, Dawn Song, Ion Stoica.</p>
<p class="content">International Conference on Learning Representations (ICLR). May, 2018.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/sqlnet.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/1711.04436"><strong>SQLNet: Generating Structured Queries From Natural Language Without Reinforcement Learning</strong></a></p>
<p class="content">Xiaojun Xu, Chang Liu, Dawn Song.</p>
<p class="content">November, 2017.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/recursion.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/1704.06611"><strong>Making Neural Programming Architectures Generalize via Recursion</strong></a></p>
<p class="content">Jonathon Cai, Richard Shin, Dawn Song.</p>
<p class="content">International Conference on Learning Representations (ICLR). April, 2017. <em><strong><font color="red">Best Paper Award</font></strong></em></p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<table border="0" cellpadding="0" cellspacing="15" width="100%">
<tbody><tr>
<td width="200"><img src="imgs/latent-attention.png" border="1"width="210"></a></td>
<td width="20"></td>
<td valign="middle" width="800"><p class="content"><a href="https://arxiv.org/abs/1611.01867"><strong>Latent Attention For If-Then Program Synthesis</strong></a></p>
<p class="content">Xinyun Chen, Chang Liu, Richard Shin, Dawn Song, Mingcheng Chen.</p>
<p class="content">Advances in Neural Information Processing Systems (NIPS). December, 2016.</p>
<p class="margin-small"> </p>
</tr>
</tbody></table>
<hr>
<h2 class="label"><a name="members"><span >Members</span></a></h2>
<ul>
<li>
<p class="content"><b>Faculty:</b> <a href="http://www.cs.berkeley.edu/~dawnsong/">Dawn Song</a></p>
</li><br>
<li><p class="content"><b>Postdocs:</b></p><ul>
<li><p class="content"><a href="http://liuchang.co/">Chang Liu</a></p></li>
</ul><br></li>
<li><p class="content"><b>Ph.D. Students:</b></p>
<ul>
<li><p class="content"><a href="https://rshin.github.io/">Richard Shin</a></p></li>
<li><p class="content"><a href="https://jungyhuk.github.io/">Xinyun Chen</p></li>
</ul>
<br>
</li>
</li>
</ul>
<br><br></body></html>