1<!DOCTYPE HTML> 2<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8"> 3 4 <title>Jiayi Chen</title> 5 6 <meta name="author" content="Jiayi Chen"> 7 <meta name="viewport" content="width=device-width, initial-scale=1"> 8 9 <link rel="stylesheet" type="text/css" href="stylesheet.css"> 10</head> 11 12<body> 13 <table style="width:100%;max-width:800px;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody> 14 <tr style="padding:0px"> 15 <td style="padding:0px"> 16 <table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody> 17 <tr style="padding:0px"> 18 <td style="padding:2.5%;width:63%;vertical-align:middle"> 19 <p class="name" style="text-align: center;"> 20 Jiayi Chen | éåæ¯ 21 </p> 22 <p>I am a final-year Ph.D. student in computer science at Peking University, advised by Prof. <a href="https://hughw19.github.io/">He Wang</a>. 23 </p> 24 <p> 25 My research focuses on <strong>low-level robot planning and control</strong>, particularly contact-rich dexterous grasping and manipulation. I believe model-based planning and control offer an efficient and generalizable way to incorporate physical priors into scalable data generation for robot learning. I am actively exploring this direction in my research. 26 </p> 27 <p style="text-align:center"> 28 <a href="mailto:[email protected]">Email</a>  /  29 <a href="https://scholar.google.com/citations?user=2lmpbXIAAAAJ&hl=zh-CN&oi=ao">Google Scholar</a>  /  30 <a href="https://github.com/jychen18/">Github</a> 31 </p> 32 </td> 33 <td style="padding:2.5%;width:40%;max-width:40%"> 34 <a href="images/jiayi.jpg"><img style="width:100%;max-width:100%" alt="profile photo" src="images/jiayi.jpg" class="hoverZoomLink"></a> 35 </td> 36 </tr> 37 </tbody></table> 38 <table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody> 39 <tr> 40 <td style="padding:20px;width:100%;vertical-align:middle"> 41 <h2>Selected Publications</h2> 42 43 <p>First-author and co-first-author papers. <a href="https://scholar.google.com/citations?user=2lmpbXIAAAAJ&hl=zh-CN&oi=ao">Full publication list on Google Scholar</a>.</p> 44 <p>Highlighted publications are shown with a shaded background.</p> 45 <p>*: equal contribution; â : corresponding author(s)</p> 46 </td> 47 </tr> 48 </tbody></table> 49 <table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody> 50 51 <tr bgcolor="#ffffdc"> 52 <td style="padding:10px;width:30%;vertical-align:middle"> 53 <img src="images/KPGrasp.png" width="100%" alt="KPGrasp overview"> 54 </td> 55 <td style="padding:20px;width:70%;vertical-align:middle"> 56 <a href="https://arxiv.org/abs/2606.09314"> 57 <span class="papertitle">KPGrasp: Scalable Keypoint Flow Matching for Dexterous Grasp Generation</span> 58 </a> 59 <br> 60 <a href="https://yuansen-huang.github.io/">Yuansen Huang*</a>, 61 <strong>Jiayi Chen</strong>*, 62 <a href="https://lhrrhl0419.github.io/">Haoran Liu</a>, 63 <a href="https://gfrl.github.io/">Yubin Ke</a>, 64 <a href="https://xjtbinghan.github.io/homepage.github.io/">Bing Han</a>, 65 <a href="https://jiangranlv.github.io/">Jiangran Lyu</a>, 66 <a href="https://miyandoris.github.io/">Mi Yan</a>, 67 <a href="https://ericyi.github.io/">Li Yi</a>, 68 <a href="https://hughw19.github.io/">He Wangâ </a> 69 <br> 70 <em>CoRL</em>, 2026 71 <br> 72 <a href="https://arxiv.org/abs/2606.09314">arXiv</a> 73 <p></p> 74 </td> 75 </tr> 76 77 <tr> 78 <td style="padding:10px;width:30%;vertical-align:middle"> 79 <img src="images/GraspADMM.png" width="100%" alt="GraspADMM overview"> 80 </td> 81 <td style="padding:20px;width:70%;vertical-align:middle"> 82 <a href="https://github.com/JYChen18/Dexonomy/blob/main/getting_started/3_ADMM.md">
83 <span class="papertitle">GraspADMM: Improving Dexterous Grasp Synthesis via ADMM Optimization</span> 84 </a> 85 <br> 86 <a href="https://lwruan.com/">Liangwang Ruan*</a>, <strong>Jiayi Chen</strong>*, <a href="https://hughw19.github.io/">He Wangâ </a>, <a href="https://cfcs.pku.edu.cn/baoquan">Baoquan Chenâ </a> 87 <br> 88 <em>IROS</em>, 2026 89 <br> 90 <a href="https://arxiv.org/abs/2603.13832">arXiv</a> / <a href="https://github.com/JYChen18/Dexonomy/blob/main/getting_started/3_ADMM.md">code</a> 91 <p></p> 92 </td> 93 </tr> 94 95 <tr> 96 <td style="padding:10px;width:30%;vertical-align:middle"> 97 <img src="images/TacDexGrasp.png" width="100%" alt="TacDexGrasp overview"> 98 </td> 99 <td style="padding:20px;width:70%;vertical-align:middle"> 100 <a href="https://arxiv.org/abs/2603.07040"> 101 <span class="papertitle">TacDexGrasp: Compliant and Robust Dexterous Grasping with Tactile Feedback</span> 102 </a> 103 <br> 104 <a href="https://gfrl.github.io/">Yubin Ke*</a>, <strong>Jiayi Chen</strong>*, Hang Lv*, Xiao Zhou, <a href="https://hughw19.github.io/">He Wangâ </a> 105 <br> 106 <em>arXiv preprint</em>, 2026 107 <br> 108 <a href="https://arxiv.org/abs/2603.07040">arXiv</a> 109 <p></p> 110 </td> 111 </tr> 112 113 <tr> 114 <td style="padding:10px;width:30%;vertical-align:middle"> 115 <img src="images/DiffCollision.png" width="100%" alt="DiffCollision overview"> 116 </td> 117 <td style="padding:20px;width:70%;vertical-align:middle"> 118 <a href="https://github.com/JYChen18/DiffCollision"> 119 <span class="papertitle">Robust Differentiable Collision Detection for General Objects</span> 120 </a> 121 <br> 122 <strong>Jiayi Chen</strong>, <a href="https://scholar.google.com/citations?hl=zh-CN&user=ZTpRlBcAAAAJ">Wei Zhao</a>, <a href="https://lwruan.com/">Liangwang Ruan</a>, <a href="https://cfcs.pku.edu.cn/baoquan">Baoquan Chen</a>, <a href="https://hughw19.github.io/">He Wangâ </a> 123 <br> 124 <em>ICRA</em>, 2026 125 <br> 126 <a href="https://arxiv.org/abs/2511.06267">arXiv</a> / <a href="https://github.com/JYChen18/DiffCollision">code</a> 127 <p></p> 128 </td> 129 </tr> 130 131 <tr bgcolor="#ffffdc"> 132 <td style="padding:10px;width:30%;vertical-align:middle"> 133 <img src='images/Dexonomy.png' width="100%"> 134 </td> 135 <td style="padding:20px;width:70%;vertical-align:middle"> 136 <a href="https://pku-epic.github.io/Dexonomy/"> 137 <span class="papertitle">Dexonomy: Synthesizing All Dexterous Grasp Types in a Grasp Taxonomy</span> 138 </a> 139 <br> 140 <strong>Jiayi Chen</strong>*, 141 <a href="https://gfrl.github.io/">Yubin Ke*</a>, 142 Lin Peng, 143 <a href="https://hughw19.github.io/">He Wangâ </a> 144 <br> 145 <em>RSS</em>, 2025 146 <br> 147 <a href="https://pku-epic.github.io/Dexonomy/">project page</a> / 148 <a href="https://arxiv.org/abs/2504.18829">arXiv</a> / 149 <a href="https://github.com/JYChen18/Dexonomy">
149code</a> / 150 <p></p> 151 </td> 152 </tr> 153 154 155 <tr> 156 <td style="padding:10px;width:30%;vertical-align:middle"> 157 <img src='images/BODex.png' width="100%"> 158 </td> 159 <td style="padding:20px;width:70%;vertical-align:middle"> 160 <a href="https://pku-epic.github.io/BODex/"> 161 <span class="papertitle">BODex: Scalable and Efficient Robotic Dexterous Grasp Synthesis Using Bilevel Optimization</span> 162 </a> 163 <br> 164 <strong>Jiayi Chen</strong>*, 165 <a href="https://gfrl.github.io/">Yubin Ke*</a>, 166 <a href="https://hughw19.github.io/">He Wangâ </a> 167 <br> 168 <em>ICRA</em>, 2025 169 <br> 170 <a href="https://pku-epic.github.io/BODex/">project page</a> / 171 <a href="https://arxiv.org/abs/2412.16490">arXiv</a> / 172 <a href="https://github.com/JYChen18/BODex">code</a> / 173 <a href="https://github.com/JYChen18/DexGraspBench">benchmark</a> / 174 <a href="https://huggingface.co/datasets/JiayiChenPKU/BODex">dataset</a> 175 <p></p> 176 </td> 177 </tr> 178 179 <tr> 180 <td style="padding:10px;width:30%;vertical-align:middle"> 181 <img src='images/TaskDexGrasp.png' width="100%"> 182 </td> 183 <td style="padding:20px;width:70%;vertical-align:middle"> 184 <a href="https://pku-epic.github.io/TaskDexGrasp/"> 185 <span class="papertitle"> 186 Task-Oriented Dexterous Hand Pose Synthesis Using Differentiable Grasp Wrench Boundary Estimator 187 </span> 188 </a> 189 <br> 190 <strong>Jiayi Chen</strong>, 191 <a href="https://chen01yx.github.io/">Yuxing Chen</a>, 192 <a href="https://mzhmxzh.github.io/">Jialiang Zhang</a>, 193 <a href="https://hughw19.github.io/">He Wangâ </a> 194 <br> 195 <em>IROS</em>, 2024 196 <br> 197 <a href="https://pku-epic.github.io/TaskDexGrasp/">project page</a> / 198 <a href="https://arxiv.org/abs/2309.13586">arXiv</a> / 199 <a href="https://github.com/JYChen18/TaskDexGrasp">code</a> 200 <p></p> 201 </td> 202 </tr> 203 204 <tr> 205 <td style="padding:10px;width:30%;vertical-align:middle"> 206 <img src='images/HOTrack.gif' width="100%"> 207 </td> 208 <td style="padding:20px;width:70%;vertical-align:middle"> 209 <a href="https://pku-epic.github.io/HOtrack/"> 210 <span class="papertitle"> 211 Tracking and Reconstructing Hand Object Interactions from Point Cloud Sequences in the Wild 212 </span> 213 </a> 214 <br> 215 <strong>Jiayi Chen*</strong>, 216 <a href="https://miyandoris.github.io/">Mi Yan*</a>, 217 <a href="https://jzhzhang.github.io/">Jiazhao Zhang</a>, 218 <a href="https://xyz-99.github.io/">Yinzhen Xu</a>, 219 <a href="https://dragonlong.github.io/"> Xiaolong Li</a>, 220 <a href="https://yijiaweng.github.io/">Yijia Weng</a>, 221 <a href="https://ericyi.github.io/">Li Yi</a>, 222 <a href="https://www.cs.columbia.edu/~shurans/">Shuran Song</a>, 223 <a href="https://hughw19.github.io/">He Wangâ </a> 224 <br> 225 <em>AAAI</em>, 2023   <strong>(Oral)</strong> 226 <br> 227 <a href="https://pku-epic.github.io/HOtrack/">project page</a> / 228 <a href="https://arxiv.org/abs/2209.12009">arXiv</a> / 229 <a href="https://github.com/PKU-EPIC/HOTrack">code</a> 230 <p></p> 231 </td> 232 </tr> 233 234 <tr> 235 <td style="padding:10px;width:30%;vertical-align:middle"> 236 <img src='images/rpmg.png' width="100%"> 237 </td> 238 <td style="padding:20px;width:70%;vertical-align:middle"> 239 <a href="https://jychen18.github.io/RPMG/">
240 <span class="papertitle"> 241 Projective Manifold Gradient Layer for Deep Rotation Regression 242 </span> 243 </a> 244 <br> 245 <strong>Jiayi Chen</strong>, 246 <a href="https://yd-yin.github.io">Yingda Yin</a>, 247 <a href="https://tolgabirdal.github.io/">Tolga Birdal</a>, 248 <a href="https://cfcs.pku.edu.cn/baoquan">Baoquan Chen</a>, 249 <a href="https://profiles.stanford.edu/leonidas-guibas">Leonidas Guibas</a>, 250 <a href="https://hughw19.github.io/">He Wangâ </a> 251 <br> 252 <em>CVPR</em>, 2022 253 <br> 254 <a href="https://jychen18.github.io/RPMG/">project page</a> / 255 <a href="https://arxiv.org/abs/2110.11657">arXiv</a> / 256 <a href="https://github.com/jychen18/RPMG">code</a> 257 <p></p> 258 </td> 259 </tr> 260 261 </tbody></table> 262 263 <table width="100%" align="center" border="0" cellspacing="0" cellpadding="20"><tbody> 264 <tr> 265 <td> 266 <h2>Experience</h2> 267 </td> 268 </tr> 269 </tbody></table> 270 <table width="100%" align="center" border="0" cellpadding="10"><tbody> 271 <tr> 272 <td style="padding-left:20px;padding-right:20px;width:20%;vertical-align:middle"><img src="images/peking_university.png", width="90%"></td> 273 <td width="80%" valign="center"> 274 <b>Peking University</b>, China 275 <br> 2022.09 - Present 276 <br><strong>Ph.D. Student in Computer Science</strong> 277 <br> Advisor: Prof. <a href="https://hughw19.github.io/">He Wang</a> 278 </td> 279 </tr> 280 <tr> 281 <td style="padding-left:20px;padding-right:20px;width:20%;vertical-align:middle"><img src="images/peking_university.png", width="90%"></td> 282 <td width="80%" valign="center"> 283 <b>Peking University</b>, China 284 <br> 2018.09 - 2022.07 285 <br><strong>B.S. in Computer Science</strong> 286 <br> Advisor: Prof. <a href="https://hughw19.github.io/">He Wang</a> 287 </td> 288 </tr> 289 290 </tbody></table> 291 292 293 <table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody> 294 <tr> 295 <td style="padding:0px"> 296 <br> 297 <p style="text-align:right;font-size:small;"> 298 Template stolen from <a href="https://jonbarron.info/">Jon Barron</a>. 299 <br> Last updated: September 20, 2026 300 </p> 301 </td> 302 </tr> 303 </tbody></table> 304 </td> 305 </tr> 306 </table> 307</body> 308 309</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.