PageSourceSearch

https://yunfeixie233.github.io/

html yunfeixie233.github.io collected 2026-10-03 10:13:42 UTC 25,516 bytes, 315 lines download raw bytes

1<!DOCTYPE HTML>
2<html lang="en">
3
4<head>
5  <meta charset="UTF-8">
6  <title>Yunfei Xie</title>
7  <meta name="author" content="Yunfei Xie">
8  <meta name="description" content="Yunfei Xie, CS PhD student at Rice University. Research on robot learning from human video and post-training for foundation models.">
9  <meta name="viewport" content="width=device-width, initial-scale=1">
10  <link rel="shortcut icon" href="images/rice.png" type="image/x-icon">
11  <link rel="stylesheet" type="text/css" href="stylesheet.css">
12</head>
13
14<body>
15  <table class="page" style="width:100%;max-width:800px;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
16    <tbody>
17      <tr style="padding:0px">
18        <td style="padding:0px">
19          <table class="profile" style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
20            <tbody>
21              <tr style="padding:0px">
22                <td style="padding:2.5%;width:60%;vertical-align:middle">
23                  <p class="name" style="text-align: center;">
24                    Yunfei Xie | 谢云飞
25                  </p>
26                  <p>
27                    I'm a CS PhD student in the Department of Computer Science at
28                    <a href="https://www.rice.edu/">Rice University</a>, advised by
29                    <a href="https://weichen582.github.io/">Prof.&nbsp;Chen Wei</a>.
30                    My research focuses on robot learning from human video and post-training for foundation models.
31                  </p>
32                  <p>
33                    I completed my bachelor's degree in the
34                    School of Artificial Intelligence and Automation
35                    at <a href="https://english.hust.edu.cn/">Huazhong University of Science &amp; Technology</a>.
36                    Previously, I was a research intern at <a href="https://research.adobe.com/">Adobe Research</a>, advised by <a href="https://bkveton.com/">Dr. Branislav Kveton</a>, at <a href="https://ucsc-vlaa.github.io/">VLAA, UC Santa Cruz</a>, advised by <a href="https://yuyinzhou.github.io/">Prof. Yuyin Zhou</a> and <a href="https://cihangxie.github.io/">Prof. Cihang Xie</a>, and at <a href="https://ccvl.jhu.edu/">CCVL, Johns Hopkins University</a>, advised by <a href="https://www.cs.jhu.edu/~ayuille/">Prof. Alan Yuille</a>.
37                  </p>
38                  <p style="font-weight:bold;">
39                    I am open for collaborations in research.
40                  </p>
41                  <p style="text-align:center">
42                    <a href="mailto:[email protected]">Email</a> &nbsp;/&nbsp;
43                    <a href="https://scholar.google.com/citations?user=rD8RC-EAAAAJ&hl=en">Google Scholar</a> &nbsp;/&nbsp;
44                    <a href="https://github.com/yunfeixie233">Github</a> &nbsp;/&nbsp;
45                    <a href="https://x.com/xiynfi1520580">Twitter</a> &nbsp;/&nbsp;
46                    <a href="https://www.linkedin.com/in/yunfei-xie-91558726a/">LinkedIn</a>
47                  </p>
48                </td>
49                <td style="padding:2.5%;width:40%;max-width:40%">
50                  <div style="position:relative;display:inline-block;">
51                    <a href="images/me.jpg"><img style="width:100%;max-width:100%;object-fit: cover; border-radius: 50%; height: 270px;" alt="profile photo" src="images/me.jpg" class="hoverZoomLink"></a>
52                    <span style="position:absolute;bottom:0;left:50%;transform:translateX(-50%);background:rgba(0,0,0,0.6);color:#fff;font-size:11px;padding:2px 8px;border-radius:10px;opacity:0;transition:opacity 0.3s;pointer-events:none;white-space:nowrap;">Photo credit: my partner Xuan</span>
53                  </div>
54                  <style>td div:hover > span { opacity: 1 !important; }</style>
55                </td>
56              </tr>
57            </tbody>
58          </table>
59          <h2>News</h2>
60          <ul style="list-style-type: disc; padding-left: 20px; margin-bottom: 20px;">
61            <li style="margin-bottom: 8px;"><strong>Sep. 2026:</strong> New work on robot learning: <a href="https://arxiv.org/abs/2609.03199">RoboTok</a> and <a href="https://yunfeixie233.github.io/s2/">Sensorimotor Alignment (S²)</a>.</li>
62            <li style="margin-bottom: 8px;"><strong>May 2026:</strong> <a href="https://yunfeixie233.github.io/MEMO/" target="_blank">MEMO</a> is accepted to ICML 2026!</li>
63            <li style="margin-bottom: 8px;"><strong>Apr. 2026:</strong> Excited to start a research internship at <a href="https://research.adobe.com/" target="_blank">Adobe Research</a> this summer!</li>
64            <li style="margin-bottom: 8px;"><strong>Feb. 2026:</strong> <a href="https://yunfeixie233.github.io/ViGaL/" target="_blank">Play to Generalize</a> and <a href="https://jwmao1.github.io/storyiter/" target="_blank">Story-Iter</a> are accepted to ICLR 2026!</li>
65            <li style="margin-bottom: 8px;"><strong>Dec. 2025:</strong> Presented <a href="https://yunfeixie233.github.io/ViGaL/" target="_blank">Play to Generalize</a> at NeurIPS '25 at the <a href="https://luma.com/3wy9hhmm" target="_blank">Open AGI Symposium</a> hosted by <a href="https://www.sentient.xyz/" target="_blank">Sentient</a>.</li>
66            <li style="margin-bottom: 8px;"><strong>Sep. 2025:</strong>
66 Started my Ph.D. journey in Computer Science at <a href="https://www.rice.edu/" target="_blank">Rice University</a>, working with <a href="https://weichen582.github.io/" target="_blank">Prof. Chen Wei</a>!</li>
67            <li style="margin-bottom: 8px;"><strong>Aug. 2025:</strong> Received the <a href="https://lambda.ai/research" target="_blank">Lambda Research Grant</a> ($10,000). Thank you, Lambda!</li>
68            <li style="margin-bottom: 8px;"><strong>Jan. 2025:</strong> <a href="https://yunfeixie233.github.io/MedTrinity-25M/" target="_blank">MedTrinity-25M</a> is accepted to ICLR 2025!</li>
69            <li style="margin-bottom: 8px;"><strong>Jul. 2024:</strong> <a href="https://arxiv.org/abs/2409.01353" target="_blank">From Pixels to Objects</a> is accepted to ECCV 2024!</li>
70          </ul>
71
72          <h2>Selected Publications</h2>
73          <p>For the full publication list, see <a href="https://scholar.google.com/citations?user=rD8RC-EAAAAJ&amp;hl=en">Google Scholar</a>.</p>
74          <table class="publications">
75            <tbody>
76              <tr>
77                <td class="publication-image">
78                  <a href="https://arxiv.org/abs/2609.03199"><img src="images/robotok.png" alt="RoboTok human manipulation videos and reconstructed hand trajectories" loading="lazy"></a>
79                </td>
80                <td class="publication-details">
81                  <a href="https://arxiv.org/abs/2609.03199"><span class="papertitle">RoboTok: An Internet-Scale Data Engine for Human Demonstration Retrieval and Dexterous Manipulation Learning</span></a><br>
82                  <span class="publication-authors"><a href="https://howardqian201.github.io/">Howard H. Qian</a>, <a href="https://www.chen-yiting.com/">Yiting Chen</a>, <strong>Yunfei Xie</strong>, <a href="https://charlierkj.github.io/">Kejia Ren</a>, <a href="https://podshara.github.io/">Podshara Chanrungmaneekul</a>, <a href="https://vector-wangel.github.io/">Gaotian Wang</a>, <a href="https://wenbowen123.github.io/">Bowen Wen</a>, <a href="https://weichen582.github.io/">Chen Wei</a>, <a href="https://hangkaiyu.github.io/">Kaiyu Hang</a></span><br>
83                  <span class="publication-venue">Preprint (under review), 2026</span><br>
84                  <a href="https://arxiv.org/abs/2609.03199">paper</a> / <a href="https://rice-robotpi-lab.github.io/RoboTok/">website</a>
85                  <p><strong>TL;DR:</strong> We retrieve manipulation-relevant human demonstrations from 100,000 web video clips using 3D hand-motion trajectories, improving dexterous-hand RL policies in Isaac Gym.</p>
86                </td>
87              </tr>
88              <tr>
89                <td class="publication-image">
90                  <a href="https://yunfeixie233.github.io/s2/"><img src="images/s2.png" alt="Sensorimotor Alignment connects visual observations, demonstrated motion, and policy success" loading="lazy"></a>
91                </td>
92                <td class="publication-details">
93                  <a href="https://yunfeixie233.github.io/s2/"><span class="papertitle">Sensorimotor Alignment: A Training-Free Predictor of Robot Policy Success</span></a><br>
94                  <span class="publication-authors"><strong>Yunfei Xie</strong>, <a href="https://howardqian201.github.io/">Howard H. Qian</a>, <a href="https://voidrank.github.io/">Shiyi Lan</a>, <a href="https://hangkaiyu.github.io/">Kaiyu Hang</a>, <a href="https://weichen582.github.io/">Chen Wei</a></span><br>
95                  <span class="publication-venue">Preprint (under review), 2026</span><br>
96                  <a href="https://yunfeixie233.github.io/s2/">website</a>
97                  <p><strong>TL;DR:</strong> We compare model representations with recorded motion trajectories to assess robot-policy and backbone capability using a training-free offline score.</p>
98                </td>
99              </tr>
100              <tr>
101                <td class="publication-image">
102                  <a href="https://yunfeixie233.github.io/ViGaL/"><img src="images/vigal.png" alt="Visual Game Learning method overview" loading="lazy"></a>
103                </td>
104                <td class="publication-details">
105                  <a href="https://yunfeixie233.github.io/ViGaL/"><span class="papertitle">Play to Generalize: Learning to Reason Through Game Play</span></a><br>
106                  <span class="publication-authors"><strong>Yunfei Xie</strong>, <a href="https://openreview.net/profile?id=~Yinsong_Ma1">Yinsong Ma</a>, <a href="https://voidrank.github.io/">Shiyi Lan</a>, <a href="https://www.cs.jhu.edu/~ayuille/">Alan Yuille</a>, <a href="https://lambert-x.github.io/">Junfei Xiao</a>, <a href="https://weichen582.github.io/">Chen Wei</a></span><br>
107                  <span class="publication-venue">ICLR 2026</span><br>
108                  <a href="https://yunfeixie233.github.io/ViGaL/">website</a> / <a href="https://arxiv.org/abs/2506.08011">paper</a>
109                  <p><strong>TL;DR:</strong> We show that RL post-training on simple arcade games improves multimodal LLMs’ out-of-domain math and spatial reasoning without worked solutions or diagrams during that post-training stage.</p>
110                </td>
111              </tr>
112              <tr>
113                <td class="publication-image">
114                  <a href="https://arxiv.org/abs/2603.09022"><img src="images/memo.png" alt="MEMO context optimization framework" loading="lazy"></a>
115                </td>
116                <td class="publication-details">
117                  <a href="https://arxiv.org/abs/2603.09022"><span class="papertitle">MEMO: Memory-Augmented Model Context Optimization for Robust Multi-Turn Multi-Agent LLM Games</span></a><br>
118                  <span class="publication-authors"><strong>Yunfei Xie</strong>, <a href="https://www.kevin-ai.com/">Kevin Wang</a>, <a href="https://x.com/bobbycxy">Bobby Cheng</a>, <a href="https://scholar.google.com/citations?user=L5XDYTIAAAAJ&amp;hl=en">Jianzhu Yao</a>, <a href="https://jamessand.github.io/">Zhizhou Sha</a>, <a href="https://www.linkedin.com/in/alex-d/">Alexander Duffy</a>, <a href="https://scholar.google.com/citations?user=34s2YS0AAAAJ&amp;hl=en">Yihan Xi</a>, <a href="https://www.hongyuanmei.com/">Hongyuan Mei</a>, <a href="https://scholar.google.com/citations?user=Up0UYEYAAAAJ&amp;hl=en">Cheston Tan</a>, <a href="https://weichen582.github.io/">Chen Wei</a>, <a href="https://scholar.google.com/citations?user=lPycXNcAAAAJ&amp;hl=en">Pramod Viswanath</a>, <a href="https://scholar.google.com/citations?user=pxFyKAIAAAAJ&amp;hl=en">Zhangyang Wang</a></span><br>
119                  <span class="publication-venue">ICML 2026</span><br>
120                  <a href="https://arxiv.org/abs/2603.09022">paper</a> / <a href="https://yunfeixie233.github.io/MEMO/">website</a> / <a href="https://github.com/openverse-ai/MEMO">code</a>
121                  <p><strong>TL;DR:</strong> Scores in multi-turn, multi-agent LLM games swing widely from run to run. MEMO treats the model’s context as something to optimize during self-play — a persistent memory bank that distills past games into reusable insights, plus TrueSkill-guided prompt evolution — roughly doubling win rates across five text games while making rankings far more stable.</p>
122                </td>
123              </tr>
124              <tr>
125                <td class="publication-image">
126                  <a href="https://benjamint2048.github.io/mu-grpo-project-page/"><img src="images/mu-grpo.png" alt="Training time comparison for GRPO and µ-GRPO" loading="lazy"></a>
127                </td>
128                <td class="publication-details">
129                  <a href="https://benjamint2048.github.io/mu-grpo-project-page/"><span class="papertitle">How Off-Policy Can GRPO Be? µ-GRPO for Efficient LLM Reinforcement Learning</span></a><br>
130                  <span class="publication-authors"><a href="https://github.com/BenjaminT2048">Minghao Tian</a>, <strong>Yunfei Xie</strong>, <a href="https://weichen582.github.io/">Chen Wei</a></span><br>
131                  <span class="publication-venue">Preprint (under review), 2026</span><br>
132                  <a href="https://benjamint2048.github.io/mu-grpo-project-page/">website</a> / <a href="https://arxiv.org/abs/2605.17570">paper</a>
133                  <p><strong>TL;DR:</strong> We reuse stale rollouts in a few large generation and optimization stages, matching standard GRPO’s reasoning performance at lower training cost.</p>
134                </td>
135              </tr>
136              <tr>
137                <td class="publication-image">
138                  <a href="https://yunfeixie233.github.io/MedTrinity-25M/"><img src="images/e209a4d74e8a7183bc6b5987c789aa6.png" alt="MedTrinity-25M multimodal data pipeline" loading="lazy"></a>
139                </td>
140                <td class="publication-details">
141                  <a href="https://yunfeixie233.github.io/MedTrinity-25M/"><span class="papertitle">MedTrinity-25M: A Large-scale Multimodal Dataset with Multigranular Annotations for Medicine</span></a><br>
142                  <span class="publication-authors"><strong>Yunfei Xie</strong>, <a href="https://github.com/Skylight-Lark">Ce Zhou</a>, <a href="https://github.com/HeartyHaven">Lang Gao</a>, <a href="https://chtholly17.github.io/">Juncheng Wu</a>, <a href="https://xhl-video.github.io/xianhangli/">Xianhang Li</a>, <a href="https://zhouhy.org/">Hong-Yu Zhou</a>, <a href="https://shengliu66.github.io/">Sheng Liu</a>, <a href="https://profiles.stanford.edu/lei-xing">Lei Xing</a>, <a href="https://www.james-zou.com/">James Zou</a>, <a href="https://cihangxie.github.io/">Cihang Xie</a>, <a href="https://yuyinzhou.github.io/">Yuyin Zhou</a></span><br>
143                  <span class="publication-venue">ICLR 2025. (400+ GitHub stars)</span><br>
144                  <a href="https://yunfeixie233.github.io/MedTrinity-25M/">website</a> / <a href="https://arxiv.org/abs/2408.02900">paper</a>
145                  <p><strong>TL;DR:</strong> We build an automated grounding, annotation, and filtering pipeline for multimodal medical pretraining, covering 25 million images across 10 modalities.</p>
146                </td>
147              </tr>
148              <tr>
149                <td class="publication-image">
150                  <a href="https://arxiv.org/abs/2409.01353"><img src="images/xie2024lgformer.png" alt="Hierarchical part and object segmentation" loading="lazy"></a>
151                </td>
152                <td class="publication-details">
153                  <a href="https://arxiv.org/abs/2409.01353"><span class="papertitle">From Pixels to Objects: A Hierarchical Approach for Part and Object Segmentation Using Local and Global Aggregation</span></a><br>
154                  <span class="publication-authors"><strong>Yunfei Xie</strong>, <a href="https://cihangxie.github.io/">Cihang Xie</a>, <a href="https://www.cs.jhu.edu/~ayuille/">Alan Yuille</a>, <a href="https://meijieru.com/">Jieru Mei</a></span><br>
155                  <span class="publication-venue">ECCV 2024</span><br>
156                  <a href="https://arxiv.org/abs/2409.01353">paper</a>
157                  <p><strong>TL;DR:</strong> We use local and global aggregation in a hierarchical vision transformer to improve part and object segmentation.</p>
158                </td>
159              </tr>
160              <tr>
161                <td class="publication-image">
162                  <a href="https://jwmao1.github.io/storyiter/"><img src="images/story_adapter.png" alt="Story-Iter long-story visualization examples" loading="lazy"></a>
163                </td>
164                <td class="publication-details">
165                  <a href="https://jwmao1.github.io/storyiter/"><span class="papertitle">Story-Iter: A Training-free Iterative Paradigm for Long Story Visualization</span></a><br>
166                  <span class="publication-authors"><a href="https://github.com/jwmao1">Jiawei Mao</a>, <a href="https://xk-huang.github.io/">Xiaoke Huang</a>, <strong>Yunfei Xie</strong>, Yuanqi Chang, <a href="https://thefllood.github.io/mudehui.github.io/">Mude Hui</a>, <a href="https://scholar.google.com/citations?user=JHTNigYAAAAJ&amp;hl=en">Bingjie Xu</a>, <a href="https://zheng80.github.io/">Zeyu Zheng</a>, <a href="https://scholar.google.com/citations?user=GgD-B68AAAAJ&amp;hl=en">Zirui Wang</a>, <a href="https://cihangxie.github.io/">Cihang Xie</a>, <a href="https://yuyinzhou.github.io/">Yuyin Zhou</a></span><br>
167                  <span class="publication-venue">ICLR 2026. (900+ GitHub stars)</span><br>
168                  <a href="https://jwmao1.github.io/storyiter/">website</a> / <a href="https://arxiv.org/abs/2410.06244">paper</a> / <a href="https://github.com/UCSC-VLAA/story-iter">code</a>
169                  <p><strong>TL;DR:</strong> We improve consistency in long-story visualization with a training-free method that refines story images using global reference context from the previous iteration.</p>
170                </td>
171              </tr>
172            </tbody>
173          </table>
174
175          <table width="100%" align="center" border="0" cellspacing="0" cellpadding="20">
176            <tbody>
177              <tr>
178                <td>
179                  <h2>Education</h2>
180                </td>
181              </tr>
182            </tbody>
183          </table>
184
185          <table width="100%" align="center" border="0" cellpadding="20">
186            <tbody>
187              <tr>
188                <td style="padding:20px;width:25%;vertical-align:middle;text-align: right;">
189                  <img src="images/rice.png" style="width: auto; height: 99px;">
190                </td>
191                <td width="75%" valign="center">
192                  Ph.D. in Computer Science,
193                  <a href="https://www.rice.edu/" target="_blank">Rice University</a>
194                  <br>
195                  Sep. 2025 - Present, Houston, TX<br>
196                  Advisor: <a href="https://weichen582.github.io/">Prof. Chen Wei</a>
197                </td>
198              </tr>
199              <tr>
200                <td style="padding:20px;width:25%;vertical-align:middle;text-align: right;">
201                  <img src="images/hust_icon.png" style="width: 65%; height: 99px;">
202                </td>
203                <td width="75%" valign="center">
204                  B.Eng. in Artificial Intelligence,
205                  <a href="https://english.hust.edu.cn/" target="_blank">Huazhong University of Science & Technology</a>
206                  <br>
207                  Sep. 2021 - Jun. 2025, Wuhan, China
208                </td>
209              </tr>
210            </tbody>
211          </table>
212
213          <table width="100%" align="center" border="0" cellspacing="0" cellpadding="20">
214            <tbody>
215              <tr>
216                <td>
217                  <h2>Experience</h2>
218                </td>
219              </tr>
220            </tbody>
221          </table>
222
223          <table class="experience-table">
224            <tbody>
225              <tr>
226                <td class="institution-image"><a class="institution-label" href="https://research.adobe.com/">Adobe<br>Research</a></td>
227                <td class="experience-details"><p><strong>May 2026 - Aug. 2026: Research Intern, <a href="https://research.adobe.com/">Adobe Research</a></strong><br>
228                  Advisor: <a href="https://bkveton.com/">Dr. Branislav Kveton</a><br>
229                  <strong>RL for Flow Matching:</strong> Developed a reinforcement learning algorithm for post-training flow-matching generative models, applied to Adobe’s Firefly foundation model.</p></td>
230              </tr>
231              <tr>
232                <td class="institution-image">
233                  <img src="images/rice.png" alt="Rice University">
234                </td>
235                <td class="experience-details"><p><strong>Feb. 2025 - Present: Research Assistant, <a href="https://www.rice.edu/">Rice University</a></strong><br>
236                  Advisor: <a href="https://weichen582.github.io/">Prof. Chen Wei</a>; robot learning in collaboration with <a href="https://hangkaiyu.github.io/">Prof. Kaiyu Hang</a>’s <a href="https://robotpilab.github.io/">RobotPI Lab</a><br>
237                  Focus: Human-video learning and offline robot-model evaluation (RoboTok, S²), RL post-training (Play to Generalize, µ-GRPO), and self-improving agents (MEMO).</p></td>
238              </tr>
239              <tr>
240                <td class="institution-image">
241                  <img src="images/ucsc_icon.png" alt="UC Santa Cruz">
242                </td>
243                <td class="experience-details"><p><strong>Dec. 2023 - Feb. 2025: Research Intern, <a href="https://ucsc-vlaa.github.io/">VLAA Lab</a>, <a href="https://www.ucsc.edu/">University of California, Santa Cruz</a></strong><br>
244                  Advisors: <a href="https://yuyinzhou.github.io/">Prof. Yuyin Zhou</a> and <a href="https://cihangxie.github.io/">Prof. Cihang Xie</a><br>
245                  Focus: Multimodal data synthesis and annotation (MedTrinity-25M), and long-story visualization (Story-Iter).</p></td>
246              </tr>
247              <tr>
248                <td class="institution-image">
249                  <img src="images/jhu_icon.png" alt="Johns Hopkins University">
250                </td>
251                <td class="experience-details"><p><strong>Jul. 2023 - Dec. 2023: Research Intern, <a href="https://ccvl.jhu.edu/">CCVL Lab</a>, <a href="https://www.jhu.edu/">Johns Hopkins University</a></strong><br>
252                  Advisor: <a href="https://www.cs.jhu.edu/~ayuille/">Prof. Alan Yuille</a><br>
253                  Focus: Hierarchical vision transformers for part and object segmentation (From Pixels to Objects).</p></td>
254              </tr>
255            </tbody>
256          </table>
257
258          <section class="resume-section">
259            <h2>Honors and Awards</h2>
260            <ul>
261              <li><strong>Aug. 2025:</strong> <a href="https://lambda.ai/research">Lambda Research Grant</a> ($10,000)</li>
262              <li><strong>2022:</strong> Science and Technology Scholarship, Huazhong University of Science and Technology (top 2% in school)</li>
263            </ul>
264          </section>
265          <section class="resume-section">
266            <h2>Invited Talks</h2>
267            <p><strong>Dec. 2025:</strong> “Play to Generalize: Learning to Reason Through Game Play,” <a href="https://luma.com/3wy9hhmm">Open AGI Symposium</a>, NeurIPS 2025.</p>
268          </section>
269          <div style="width: 100%; background-color: #ffffff; padding: 20px 0;">
270            <table width="100%" align="center" border="0" cellspacing="0" cellpadding="10">
271              <tbody>
272                <tr>
273                  <td>
274                    <h2 style="margin-bottom: 5px;">Reviewer</h2>
275                  </td>
276                </tr>
277              </tbody>
278            </table>
279            <table width="100%" align="center" border="0" cellpadding="10">
280              <tbody>
281                <tr>
282                  <td width="75%" valign="center">
283                    <p style="margin-top: 0;">
284                      <strong>2026: ICLR, ICML, CVPR, ECCV</strong><br>
285                      <strong>2025: ICCV, ICLR, CVPR, ICML, TMM, NeurIPS, TPAMI</strong><br>
286                      <strong>2024: CVPR, ICML, IEEE ISBI</strong>
287                    </p>
288                  </td>
289                </tr>
290              </tbody>
291            </table>
292            
292<script type="text/javascript" id="mapmyvisitors" src="//mapmyvisitors.com/map.js?d=FUAUAqFMXEsXJBaQKTYhltP6zqOAmSRonLySDrjKcVo&cl=ffffff&w=a"></script>
292
293          </div>
294
295
296          <table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;">
297            <tbody>
298              <tr>
299                <td style="padding:0px">
300                  <br>
301                  <p style="text-align:right;font-size:small;">
302                    This website template was borrowed from <a href="https://jonbarron.info/">Jon Barron</a>. <br>
303                    Last updated on September 6, 2026.
304                  </p>
305                </td>
306              </tr>
307            </tbody>
308          </table>
309        </td>
310      </tr>
311    </tbody>
312  </table>
313</body>
314
315</html>

Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.