TimeMAE/index.html at main · ustc-time-series/TimeMAE · GitHub

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
<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="UTF-8">
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  <title>TimeMAE: Self-Supervised Representations of Time Series with Decoupled Masked Autoencoders</title>

  <!-- Google Scholar meta tags -->
  <meta name="citation_title" content="TimeMAE: Self-Supervised Representations of Time Series with Decoupled Masked Autoencoders">
  <meta name="citation_author" content="Cheng, Mingyue">
  <meta name="citation_author" content="Tao, Xiaoyu">
  <meta name="citation_author" content="Liu, Zhiding">
  <meta name="citation_author" content="Liu, Qi">
  <meta name="citation_author" content="Zhang, Hao">
  <meta name="citation_author" content="Zhang, Rujiao">
  <meta name="citation_author" content="Chen, Enhong">
  <meta name="citation_publication_date" content="2026/02/27">
  <meta name="citation_conference_title" content="Proceedings of the Nineteenth ACM International Conference on Web Search and Data Mining">
  <meta name="citation_arxiv_id" content="2303.00320">
  <meta name="citation_doi" content="10.1145/3773966.3778007">
  <meta name="citation_pdf_url" content="https://arxiv.org/pdf/2303.00320">
  <meta name="citation_abstract_html_url" content="https://mingyue-cheng.github.io/TimeMAE/">
  <link rel="canonical" href="https://mingyue-cheng.github.io/TimeMAE/">

  <link rel="preconnect" href="https://fonts.googleapis.com">
  <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
  <link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">

  <style>
    *, *::before, *::after { box-sizing: border-box; margin: 0; padding: 0; }

    :root {
      --accent: #2563eb;
      --accent-hover: #1d4ed8;
      --dark: #111827;
      --mid: #374151;
      --light: #6b7280;
      --border: #e5e7eb;
      --bg: #ffffff;
      --bg-code: #f9fafb;
    }

    body {
      font-family: 'Inter', system-ui, sans-serif;
      background: var(--bg);
      color: var(--dark);
      line-height: 1.65;
      font-size: 16px;
    }

    .page {
      max-width: 860px;
      margin: 0 auto;
      padding: 48px 24px 80px;
    }

    /* Venue badge */
    .venue-badge {
      display: inline-block;
      background: var(--accent);
      color: #fff;
      font-size: 13px;
      font-weight: 600;
      letter-spacing: 0.04em;
      padding: 4px 12px;
      border-radius: 20px;
      margin-bottom: 20px;
      text-transform: uppercase;
    }

    /* Title */
    h1 {
      font-size: clamp(22px, 4vw, 28px);
      font-weight: 700;
      line-height: 1.3;
      color: var(--dark);
      margin-bottom: 20px;
    }

    /* Authors */
    .authors {
      font-size: 15px;
      color: var(--mid);
      margin-bottom: 32px;
    }
    .authors span { font-weight: 500; }

    /* Action buttons */
    .buttons {
      display: flex;
      flex-wrap: wrap;
      gap: 10px;
      margin-bottom: 48px;
    }
    .btn {
      display: inline-flex;
      align-items: center;
      gap: 6px;
      padding: 9px 20px;
      border-radius: 8px;
      font-size: 14px;
      font-weight: 600;
      text-decoration: none;
      transition: background 0.15s, transform 0.1s;
      cursor: pointer;
      border: none;
    }
    .btn:active { transform: scale(0.97); }
    .btn-primary {
      background: var(--dark);
      color: #fff;
    }
    .btn-primary:hover { background: var(--mid); }
    .btn-outline {
      background: transparent;
      color: var(--dark);
      border: 1.5px solid var(--border);
    }
    .btn-outline:hover { background: var(--bg-code); }

    /* Section headings */
    h2 {
      font-size: 18px;
      font-weight: 700;
      color: var(--dark);
      margin-bottom: 14px;
      padding-bottom: 10px;
      border-bottom: 2px solid var(--border);
    }

    /* Abstract */
    .section { margin-bottom: 48px; }
    .abstract-text {
      font-size: 15.5px;
      color: var(--mid);
      line-height: 1.8;
    }

    /* BibTeX */
    .bibtex-wrap {
      position: relative;
    }
    .bibtex-code {
      display: block;
      background: var(--bg-code);
      border: 1px solid var(--border);
      border-radius: 10px;
      padding: 20px 24px;
      font-family: 'JetBrains Mono', 'Fira Code', monospace;
      font-size: 13px;
      line-height: 1.7;
      color: var(--mid);
      white-space: pre;
      overflow-x: auto;
    }
    .copy-btn {
      position: absolute;
      top: 12px;
      right: 12px;
      background: var(--dark);
      color: #fff;
      border: none;
      border-radius: 6px;
      padding: 5px 12px;
      font-size: 12px;
      font-weight: 600;
      cursor: pointer;
      transition: background 0.15s;
    }
    .copy-btn:hover { background: var(--mid); }
    .copy-btn.copied {
      background: #16a34a;
    }

    /* SVG icons */
    .icon {
      width: 16px;
      height: 16px;
      flex-shrink: 0;
    }

    /* Divider */
    hr {
      border: none;
      border-top: 1px solid var(--border);
      margin: 48px 0;
    }

    /* Footer */
    footer {
      font-size: 13px;
      color: var(--light);
      text-align: center;
      padding-top: 16px;
    }
    footer a { color: var(--light); }

    @media (max-width: 600px) {
      .page { padding: 32px 16px 60px; }
    }
  </style>
</head>
<body>
<div class="page">

  <div class="venue-badge">WSDM 2026</div>

  <h1>TimeMAE: Self-Supervised Representations of Time Series with Decoupled Masked Autoencoders</h1>

  <p class="authors">
    <span>Mingyue Cheng</span>,
    <span>Xiaoyu Tao</span>,
    <span>Zhiding Liu</span>,
    <span>Qi Liu</span>,
    <span>Hao Zhang</span>,
    <span>Rujiao Zhang</span>,
    <span>Enhong Chen</span>
  </p>

  <div class="buttons">
    <a class="btn btn-primary" href="https://arxiv.org/abs/2303.00320" target="_blank" rel="noopener">
      <!-- arXiv icon -->
      <svg class="icon" viewBox="0 0 24 24" fill="currentColor" xmlns="http://www.w3.org/2000/svg">
        <path d="M12 2L2 7l10 5 10-5-10-5zM2 17l10 5 10-5M2 12l10 5 10-5"/>
      </svg>
      arXiv
    </a>
    <a class="btn btn-primary" href="https://arxiv.org/pdf/2303.00320" target="_blank" rel="noopener">
      <!-- PDF icon -->
      <svg class="icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round">
        <path d="M14 2H6a2 2 0 0 0-2 2v16a2 2 0 0 0 2 2h12a2 2 0 0 0 2-2V8z"/>
        <polyline points="14 2 14 8 20 8"/>
        <line x1="9" y1="13" x2="15" y2="13"/>
        <line x1="9" y1="17" x2="15" y2="17"/>
        <polyline points="9 9 10 9"/>
      </svg>
      PDF
    </a>
    <a class="btn btn-outline" href="https://github.com/Mingyue-Cheng/TimeMAE" target="_blank" rel="noopener">
      <!-- GitHub icon -->
      <svg class="icon" viewBox="0 0 24 24" fill="currentColor">
        <path d="M12 2C6.477 2 2 6.484 2 12.017c0 4.425 2.865 8.18 6.839 9.504.5.092.682-.217.682-.483 0-.237-.008-.868-.013-1.703-2.782.605-3.369-1.343-3.369-1.343-.454-1.158-1.11-1.466-1.11-1.466-.908-.62.069-.608.069-.608 1.003.07 1.531 1.032 1.531 1.032.892 1.53 2.341 1.088 2.91.832.092-.647.35-1.088.636-1.338-2.22-.253-4.555-1.113-4.555-4.951 0-1.093.39-1.988 1.029-2.688-.103-.253-.446-1.272.098-2.65 0 0 .84-.27 2.75 1.026A9.564 9.564 0 0 1 12 6.844a9.59 9.59 0 0 1 2.504.337c1.909-1.296 2.747-1.027 2.747-1.027.546 1.379.202 2.398.1 2.651.64.7 1.028 1.595 1.028 2.688 0 3.848-2.339 4.695-4.566 4.943.359.309.678.92.678 1.855 0 1.338-.012 2.419-.012 2.747 0 .268.18.58.688.482A10.02 10.02 0 0 0 22 12.017C22 6.484 17.522 2 12 2z"/>
      </svg>
      Code
    </a>
    <a class="btn btn-outline" href="https://doi.org/10.1145/3773966.3778007" target="_blank" rel="noopener">
      <!-- DOI/ACM icon -->
      <svg class="icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round">
        <circle cx="12" cy="12" r="10"/>
        <line x1="2" y1="12" x2="22" y2="12"/>
        <path d="M12 2a15.3 15.3 0 0 1 4 10 15.3 15.3 0 0 1-4 10 15.3 15.3 0 0 1-4-10 15.3 15.3 0 0 1 4-10z"/>
      </svg>
      ACM DL
    </a>
    <a class="btn btn-outline" href="#bibtex" onclick="document.getElementById('bibtex').scrollIntoView({behavior:'smooth'});return false;">
      <!-- Quote icon -->
      <svg class="icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" stroke-linejoin="round">
        <path d="M3 21c3 0 7-1 7-8V5c0-1.25-.756-2.017-2-2H4c-1.25 0-2 .75-2 1.972V11c0 1.25.75 2 2 2 1 0 1 0 1 1v1c0 1-1 2-2 2s-1 .008-1 1.031V20c0 1 0 1 1 1z"/>
        <path d="M15 21c3 0 7-1 7-8V5c0-1.25-.757-2.017-2-2h-4c-1.25 0-2 .75-2 1.972V11c0 1.25.75 2 2 2h.75c0 2.25.25 4-2.75 4v3c0 1 0 1 1 1z"/>
      </svg>
      BibTeX
    </a>
  </div>

  <!-- Abstract -->
  <div class="section">
    <h2>Abstract</h2>
    <p class="abstract-text">
      Learning transferable representations from unlabeled time series is crucial for improving performance in data-scarce classification. Existing self-supervised methods often operate at the point level and rely on unidirectional encoding, leading to low semantic density and a mismatch between pre-training and downstream optimization. In this paper, we propose TimeMAE, a self-supervised framework that reformulates masked modeling for time series via semantic unit elevation and decoupled representation learning. Instead of modeling individual time steps, TimeMAE segments time series into non-overlapping sub-series to form semantically enriched units, enabling more informative masked reconstruction while reducing computational cost. To address the representation discrepancy introduced by masking, we design a decoupled masked autoencoder that separately encodes visible and masked regions, avoiding artificial masked tokens in the main encoder. To guide pre-training, we introduce two complementary objectives: masked codeword classification, which discretizes sub-series semantics via a learned tokenizer and masked representation regression, which aligns continuous representations through a momentum-updated target encoder. Extensive experiments on five datasets demonstrate that TimeMAE outperforms competitive baselines, particularly in label-scarce scenarios and transfer learning scenarios.
    </p>
  </div>

  <!-- BibTeX -->
  <div class="section" id="bibtex">
    <h2>BibTeX</h2>
    <div class="bibtex-wrap">
      <pre id="bibtex-code" class="bibtex-code">@inproceedings{cheng2026timemae,
  author    = {Cheng, Mingyue and Tao, Xiaoyu and Liu, Zhiding and
               Liu, Qi and Zhang, Hao and Zhang, Rujiao and Chen, Enhong},
  title     = {{TimeMAE}: Self-Supervised Representations of Time Series
               with Decoupled Masked Autoencoders},
  booktitle = {Proceedings of the Nineteenth {ACM} International Conference
               on Web Search and Data Mining},
  series    = {{WSDM} '26},
  year      = {2026},
  publisher = {{ACM}},
  doi       = {10.1145/3773966.3778007},
  note      = {arXiv:2303.00320}
}</pre>
      <button class="copy-btn" onclick="copyBibtex(this)">Copy</button>
    </div>
  </div>

  <hr>

  <footer>
    <p>TimeMAE &middot; WSDM 2026 &middot; <a href="https://doi.org/10.1145/3773966.3778007">10.1145/3773966.3778007</a></p>
  </footer>

</div>

<script>
  function copyBibtex(btn) {
    navigator.clipboard.writeText(document.getElementById('bibtex-code').textContent).then(function() {
      btn.textContent = 'Copied!';
      btn.classList.add('copied');
      setTimeout(function() {
        btn.textContent = 'Copy';
        btn.classList.remove('copied');
      }, 2000);
    });
  }
</script>
</body>
</html>