1<!DOCTYPE html> 2<html lang="en"> 3 <head> 4 <meta charset="utf-8"> 5 <meta name="viewport" content="width=device-width, initial-scale=1"> 6 <title>Xiangteng He</title> 7 <link rel="stylesheet" href="/styles.css?v=20260327-2"> 8
8<script defer src="/news-data.js?v=20260327-2"></script>
8 9
9<script defer src="/news-render.js?v=20260327-2"></script>
9 10</head> 11<body class="coastal-page home-page"> 12 <header class="site-header"> 13 <div class="nav-wrap"> 14 <a class="brand" href="index.html">Xiangteng He</a> 15 <nav aria-label="Main"> 16 <ul class="nav-links"> 17 <li><a href="index.html">Home</a></li> 18 <li><a href="news.html">News</a></li> 19 <li><a href="publications.html">Publications</a></li> 20 <li><a href="datasets.html">Datasets</a></li> 21 <li><a href="awards.html">Awards</a></li> 22 </ul> 23 </nav> 24 </div> 25 </header> 26 27 <main class="page"> 28 <div class="container"> 29 <section class="card hero plain-section"> 30 <div class="portrait"> 31 <img src="/hexiangteng.jpg?v=20260327-2" alt="Portrait of Xiangteng He"> 32 </div> 33 <div> 34 <h1>Xiangteng He ä½ç¸è ¾</h1> 35 <p><strong>Postdoctoral Researcher, University of British Columbia</strong></p> 36 <p><strong>Current Research Interests:</strong> Representation learning and vision foundation models, fine-grained visual and multimodal understanding, and vision-language models.</p> 37 <div class="icon-row" aria-label="Profile links"> 38 <a class="icon-link" href="mailto:[email protected]" aria-label="Email" title="Email"> 39 <svg viewBox="0 0 24 24" width="24" height="24" aria-hidden="true"> 40 <path fill="currentColor" d="M3 5h18a1 1 0 0 1 1 1v12a1 1 0 0 1-1 1H3a1 1 0 0 1-1-1V6a1 1 0 0 1 1-1Zm17 3.2-7.45 5.58a1 1 0 0 1-1.1 0L4 8.2V17h16V8.2Zm-1.5-1.2H5.5L12 11.88 18.5 7Z"/> 41 </svg> 42 </a> 43 <a class="icon-link" href="https://scholar.google.com/citations?user=37RO0eYAAAAJ&hl=En" aria-label="Google Scholar" title="Google Scholar"> 44 <svg viewBox="0 0 24 24" width="24" height="24" aria-hidden="true"> 45 <path fill="currentColor" d="m12 3 9 5-9 5-9-5 9-5Zm-5.9 7.78L12 14l5.9-3.22L21 12.5 12 17l-9-4.5 3.1-1.72Z"/> 46 <path fill="currentColor" d="M7.5 14.2h2V18h5v-3.8h2V19a1 1 0 0 1-1 1h-7a1 1 0 0 1-1-1v-4.8Z"/> 47 </svg> 48 </a> 49 <a class="icon-link" href="https://www.linkedin.com/in/xiangteng-he-0107a9390/" aria-label="LinkedIn" title="LinkedIn"> 50 <svg viewBox="0 0 24 24" width="24" height="24" aria-hidden="true"> 51 <path fill="currentColor" d="M5 3h14a2 2 0 0 1 2 2v14a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2V5a2 2 0 0 1 2-2Zm2.2 6.3a1.45 1.45 0 1 0 0-2.9 1.45 1.45 0 0 0 0 2.9ZM6 18h2.4v-7H6v7Zm6.1 0h2.4v-3.62c0-.95.18-1.88 1.36-1.88 1.16 0 1.18 1.08 1.18 1.94V18h2.4v-4.03c0-1.98-.43-3.5-2.74-3.5-1.11 0-1.85.61-2.15 1.19h-.03V11h-2.3c.03.61 0 7 0 7Z"/> 52 </svg> 53 </a> 54 </div> 55 </div> 56 </section> 57 58 <section class="academic-market" aria-label="Academic job market announcement"> 59 <p>ð¢ <strong>I am on the academic job market for 2026â2027</strong>, seeking faculty positions in computer vision, multimodal learning, and AI.</p> 60 </section> 61 62 <!-- <section class="card plain-section notice notice-cfp"> 63 <div class="notice-text"> 64 <h2><svg class="section-icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="1.8" aria-hidden="true"><path d="M3 11l18-8-8 18-2-8-8-2Z"/></svg>Call for Papers</h2> 65 <p><a href="https://medfmb.github.io/">Workshop on Medical Foundation Models and Benchmarks</a> at <a href="https://eccv.ecva.net/">ECCV 2026</a>.</p> 66 <p> 67 We welcome submissions on new ideas, methods, datasets, benchmarks, and perspectives. Previously published papers are also welcome for non-archival presentation. 68 </p> 69 70 <p class="important-dates"> 71 <strong>Important dates:</strong><br> 72 Submission deadline: July 1, 2026, 11:59 PM AoE<br> 73 Final decisions: July 15, 2026, 11:59 PM AoE 74 </p> 75 76 </div> 77 <button 78 class="notice-poster" 79 type="button" 80 aria-label="Enlarge ECCVMed poster" 81 data-lightbox-trigger 82 data-lightbox-src="/papers/ECCV 2026-Med.png" 83 data-lightbox-alt="ECCVMed poster" 84 > 85 <img src="/papers/ECCV 2026-Med.png" alt="ECCVMed"> 86 </button> 87 </section> --> 88 89 <!-- 90 <section class="card plain-section notice notice-cfp"> 91 <div class="notice-text"> 92 <h2>
92<svg class="section-icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="1.8" aria-hidden="true"><path d="M3 11l18-8-8 18-2-8-8-2Z"/></svg>Call for Papers</h2> 93 <p><a href="https://sites.google.com/view/fgvc13/home">The 13th Workshop on Fine-Grained Visual Categorization</a> at <a href="https://cvpr.thecvf.com/Conferences/2026">CVPR 2026</a>.</p> 94 <p>FGVC13 will feature two paper tracks and a nectar track. Submission details and timelines are available on the workshop website.</p> 95 <p class="notice-links"> 96 <a href="https://sites.google.com/view/fgvc13/home">Workshop</a> 97 <a href="https://sites.google.com/view/fgvc13/submission">Submission</a> 98 </p> 99 </div> 100 <a class="notice-poster" href="https://sites.google.com/view/fgvc13/home" aria-label="FGVC13 poster"> 101 <img src="/papers/FGVC13.png?v=20260327-2" alt="FGVC13 poster"> 102 </a> 103 </section> 104 --> 105 106 <section class="card plain-section module-panel"> 107 <h2 class="section-title"><svg class="section-icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="1.8" aria-hidden="true"><path d="M20 21v-2a4 4 0 0 0-4-4H8a4 4 0 0 0-4 4v2"/><circle cx="12" cy="7" r="4"/></svg>Short Bio</h2> 108 <p>Dr. Xiangteng He is a Postdoctoral Researcher at the University of British Columbia, working with Leonid Sigal. Prior to joining UBC, he was a Research Assistant Professor at Peking University and conducted research at Alibaba DAMO Academy. He received his PhD from Peking University, where his doctoral research was recognized with the CCF Outstanding Doctoral Dissertation Award and the Baidu Scholarship.</p> 109 </section> 110 111 <section class="card plain-section research-section module-panel"> 112 <h2 class="section-title"><svg class="section-icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="1.8" aria-hidden="true"><circle cx="12" cy="12" r="4"/><path d="M12 2v3M12 19v3M2 12h3M19 12h3"/></svg>Research</h2> 113 <!-- <p class="research-summary">My research focuses on fine-grained visual understanding, representation learning and vision foundation models, and multimodal learning. I am particularly interested in learning structured and transferable representations that support fine-grained perception, multimodal reasoning, and efficient adaptation across diverse domains. My recent works explore:</p> --> 114 <ul class="research-list"> 115 <li> 116 <strong>Representation learning and vision foundation models: </strong>predictive representation learning, joint-embedding models, and world models 117 <span class="research-links research-tags"><a href="https://link.springer.com/chapter/10.1007/978-3-032-36984-0_25">ECCV'26</a></span> 118 </li> 119 <li> 120 <strong>Fine-grained visual and multimodal understanding: </strong> 121 fine-grained recognition, adaptation, and structured multimodal analysis 122 <span class="research-links research-tags"><a href="https://arxiv.org/abs/2504.05662">CVPR'26</a><a href="https://dl.acm.org/doi/abs/10.1109/TIP.2024.3459788">TIP'24</a><a href="https://www.ijcai.org/proceedings/2024/0144.pdf">IJCAI'24</a><a href="https://dl.acm.org/doi/abs/10.1145/3581783.3612403">ACM MM'23</a></span> 123 </li> 124 <li> 125 <strong>Multimodal learning and vision-language models: </strong>cross-modal alignment, multimodal understanding, and generative modeling 126 <span class="research-links research-tags"><a href="https://arxiv.org/abs/2510.08510">ICLR'26</a><a href="https://openaccess.thecvf.com/content/ICCV2025/papers/Pan_ICE-Bench_A_Unified_and_Comprehensive_Benchmark_for_Image_Creating_and_ICCV_2025_paper.pdf">ICCV'25</a><a href="https://ojs.aaai.org/index.php/AAAI/article/view/27885/27795">AAAI'24</a><a href="https://openaccess.thecvf.com/content/CVPR2023/papers/Hsu_PosterLayout_A_New_Benchmark_and_Approach_for_Content-Aware_Visual-Textual_Presentation_CVPR_2023_paper.pdf">CVPR'23</a><a href="https://openaccess.thecvf.com/content/ICCV2023/papers/Pan_Scanning_Only_Once_An_End-to-end_Framework_for_Fast_Temporal_Grounding_ICCV_2023_paper.pdf">ICCV'23</a><a href="https://dl.acm.org/doi/abs/10.1145/3581783.3612405">ACM MM'23</a></span> 127 </li> 128 <li> 129 <strong>Efficient and domain-specialized multimodal learning: </strong>
129 efficient multimodal and video understanding, together with applications to medical imaging and scientific data 130 <span class="research-links research-tags"><a href="https://arxiv.org/abs/2605.14310">NeurIPS'26</a><a href="https://arxiv.org/abs/2512.02566">CVPR'26</a><a href="https://arxiv.org/abs/2607.15556">CARVE</a><a href="https://arxiv.org/abs/2511.02996">SCALE-VLP</a></span> 131 </li> 132 </ul> 133 </section> 134 135 <section class="card plain-section module-panel"> 136 <h2 class="section-title"><svg class="section-icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="1.8" aria-hidden="true"><path d="M22 10v6M2 10l10-5 10 5-10 5-10-5Z"/><path d="M6 12v5c3 2 9 2 12 0v-5"/></svg>For Students</h2> 137 <p>To junior PhD, master's, and undergraduate students: if you would like to chat about life, career planning, or research ideas related to CV and AI, feel free to email me to schedule a meeting. I dedicate 30 minutes every week for such meetings.</p> 138 </section> 139 140 <section class="card plain-section module-panel"> 141 <h2 class="section-title"><svg class="section-icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="1.8" aria-hidden="true"><path d="M19 21H5a2 2 0 0 1-2-2V5a2 2 0 0 1 2-2h11l5 5v11a2 2 0 0 1-2 2Z"/><path d="M7 3v5h8"/></svg>Recent News</h2> 142 <ul class="news-list" data-news-list="recent"></ul> 143 <p class="news-more"><a href="news.html">View all news</a></p> 144 </section> 145 </div> 146 </main> 147 148 <div class="lightbox" data-lightbox hidden> 149 <button class="lightbox-backdrop" type="button" aria-label="Close enlarged poster" data-lightbox-close></button> 150 <div class="lightbox-panel" role="dialog" aria-modal="true" aria-label="Enlarged poster"> 151 <button class="lightbox-close" type="button" aria-label="Close enlarged poster" data-lightbox-close>×</button> 152 <img class="lightbox-image" src="" alt="" data-lightbox-image> 153 </div> 154 </div> 155 156
156<script> 157 (function () { 158 const lightbox = document.querySelector("[data-lightbox]"); 159 if (!lightbox) return; 160 161 const lightboxImage = lightbox.querySelector("[data-lightbox-image]"); 162 const closeButtons = lightbox.querySelectorAll("[data-lightbox-close]"); 163 const triggers = document.querySelectorAll("[data-lightbox-trigger]"); 164 165 function closeLightbox() { 166 lightbox.hidden = true; 167 document.body.classList.remove("lightbox-open"); 168 lightboxImage.src = ""; 169 lightboxImage.alt = ""; 170 } 171 172 function openLightbox(src, alt) { 173 lightboxImage.src = src; 174 lightboxImage.alt = alt || ""; 175 lightbox.hidden = false; 176 document.body.classList.add("lightbox-open"); 177 } 178 179 triggers.forEach((trigger) => { 180 trigger.addEventListener("click", () => { 181 openLightbox(trigger.dataset.lightboxSrc, trigger.dataset.lightboxAlt); 182 }); 183 }); 184 185 closeButtons.forEach((button) => { 186 button.addEventListener("click", closeLightbox); 187 }); 188 189 document.addEventListener("keydown", (event) => { 190 if (event.key === "Escape" && !lightbox.hidden) { 191 closeLightbox(); 192 } 193 }); 194 }()); 195 </script>
195 196</body> 197</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.