|
| 1 | +<!DOCTYPE html> |
| 2 | +<html> |
| 3 | + |
| 4 | +<head> |
| 5 | + <meta charset="utf-8"> |
| 6 | + <meta name="description" content="Efficient In-Memory Inverted Indexes: Theory and Practice."> |
| 7 | + <meta name="keywords" content="Inverted Indexes, Information Retrieval, Search Engines"> |
| 8 | + <meta name="viewport" content="width=device-width, initial-scale=1"> |
| 9 | + <title>Efficient In-Memory Inverted Indexes: Theory and Practice</title> |
| 10 | + |
| 11 | + <!-- Global site tag (gtag.js) - Google Analytics --> |
| 12 | + <script async src="https://www.googletagmanager.com/gtag/js?id=G-PYVRSFMDRL"></script> |
| 13 | + <script> |
| 14 | + window.dataLayer = window.dataLayer || []; |
| 15 | + |
| 16 | + function gtag() { |
| 17 | + dataLayer.push(arguments); |
| 18 | + } |
| 19 | + |
| 20 | + gtag('js', new Date()); |
| 21 | + |
| 22 | + gtag('config', 'G-PYVRSFMDRL'); |
| 23 | + </script> |
| 24 | + |
| 25 | + <link href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro" rel="stylesheet"> |
| 26 | + |
| 27 | + <link rel="stylesheet" href="./static/css/bulma.min.css"> |
| 28 | + <link rel="stylesheet" href="./static/css/bulma-carousel.min.css"> |
| 29 | + <link rel="stylesheet" href="./static/css/bulma-slider.min.css"> |
| 30 | + <link rel="stylesheet" href="./static/css/fontawesome.all.min.css"> |
| 31 | + <link rel="stylesheet" href="https://cdn.jsdelivr.net/gh/jpswalsh/academicons@1/css/academicons.min.css"> |
| 32 | + <link rel="stylesheet" href="./static/css/index.css"> |
| 33 | + <link rel="icon" href="./static/images/favicon.svg"> |
| 34 | + <link rel="stylesheet" href="./static/css/profile.css"> |
| 35 | + <script src="https://ajax.googleapis.com/ajax/libs/jquery/3.5.1/jquery.min.js"></script> |
| 36 | + <script defer src="./static/js/fontawesome.all.min.js"></script> |
| 37 | + <script src="./static/js/bulma-carousel.min.js"></script> |
| 38 | + <script src="./static/js/bulma-slider.min.js"></script> |
| 39 | + <script src="./static/js/index.js"></script> |
| 40 | +</head> |
| 41 | + |
| 42 | +<body> |
| 43 | + |
| 44 | + <!-- <nav class="navbar" role="navigation" aria-label="main navigation"> |
| 45 | + <div class="navbar-brand"> |
| 46 | + <a role="button" class="navbar-burger" aria-label="menu" aria-expanded="false"> |
| 47 | + <span aria-hidden="true"></span> |
| 48 | + <span aria-hidden="true"></span> |
| 49 | + <span aria-hidden="true"></span> |
| 50 | + </a> |
| 51 | + </div> |
| 52 | + <div class="navbar-menu"> |
| 53 | + <div class="navbar-start" style="flex-grow: 1; justify-content: center;"> |
| 54 | + <a class="navbar-item" href="https://keunhong.com"> |
| 55 | + <span class="icon"> |
| 56 | + <i class="fas fa-home"></i> |
| 57 | + </span> |
| 58 | + </a> |
| 59 | +
|
| 60 | + <div class="navbar-item has-dropdown is-hoverable"> |
| 61 | + <a class="navbar-link"> |
| 62 | + More Research |
| 63 | + </a> |
| 64 | + <div class="navbar-dropdown"> |
| 65 | + <a class="navbar-item" href="https://hypernerf.github.io"> |
| 66 | + HyperNeRF |
| 67 | + </a> |
| 68 | + <a class="navbar-item" href="https://nerfies.github.io"> |
| 69 | + Nerfies |
| 70 | + </a> |
| 71 | + <a class="navbar-item" href="https://latentfusion.github.io"> |
| 72 | + LatentFusion |
| 73 | + </a> |
| 74 | + <a class="navbar-item" href="https://photoshape.github.io"> |
| 75 | + PhotoShape |
| 76 | + </a> |
| 77 | + </div> |
| 78 | + </div> |
| 79 | + </div> |
| 80 | +
|
| 81 | + </div> |
| 82 | + </nav> |
| 83 | + --> |
| 84 | + |
| 85 | + <section class="hero"> |
| 86 | + <div class="hero-body"> |
| 87 | + <div class="container is-max-desktop"> |
| 88 | + <div class="columns is-centered"> |
| 89 | + <div class="column has-text-centered"> |
| 90 | + <h1 class="title is-1 publication-title">Practical, Efficient, In-Memory Inverted Indexes/h1> |
| 91 | + <div class="is-size-5 publication-authors"> |
| 92 | + <span class="author-block"> |
| 93 | + <table> |
| 94 | + <tr> |
| 95 | + <td class="profile-cell"><img class="profile-img" src="static/images/joel.png" |
| 96 | + alt="Profile Jel"></td> |
| 97 | + <td class="profile-cell"><img class="profile-img" src="static/images/sean.png" |
| 98 | + alt="Profile Sean"></td> |
| 99 | + <td class="profile-cell"><img class="profile-img" |
| 100 | + src="static/images/antonio.png" alt="Profile Antonio"></td> |
| 101 | + <td class="profile-cell"><img class="profile-img" src="static/images/michal.png" |
| 102 | + alt="Profile Michal"></td> |
| 103 | + </tr> |
| 104 | + <tr> |
| 105 | + <!-- <th scope="row">TR-7</th> --> |
| 106 | + <td width="20%" style="text-align: center"><a href="https://jmmackenzie.io" |
| 107 | + style="border-radius: 50%">Joel Mackenzie</a><sup>1</sup></td> |
| 108 | + <td width="20%" style="text-align: center"><a href="https://macavaney.us" |
| 109 | + style="border-radius: 50%">Sean MacAvaney</a><sup>2</sup></td> |
| 110 | + <td width="20%" style="text-align: center"><a href="https://antoniomallia.it" |
| 111 | + style="border-radius: 50%">Antonio Mallia</a><sup>3</sup></td> |
| 112 | + <td width="20%" style="text-align: center"><a href="https://siedlaczek.me/" |
| 113 | + style="border-radius: 50%">Michal Siedlaczek</a><sup>4</sup></td> |
| 114 | + </tr> |
| 115 | + </table> |
| 116 | + </div> |
| 117 | + <div class="is-size-5 publication-authors"> |
| 118 | + <span class="author-block"><sup>1</sup>The University of Queensland,</span> |
| 119 | + <span class="author-block"><sup>2</sup>University of Glasgow,</span> |
| 120 | + <span class="author-block"><sup>3</sup>Seltz,</span> |
| 121 | + <span class="author-block"><sup>4</sup>MongoDB, Inc.</span> |
| 122 | + </div> |
| 123 | + <br /> |
| 124 | + <div class="is-size-5 publication-authors"> |
| 125 | + <b>Date and Location: ECIR 2026, Delft, Netherlands - Sunday March 30th, 09.00-12.30</b> |
| 126 | + </div> |
| 127 | + |
| 128 | + <!-- |
| 129 | + <div class="column has-text-centered"> |
| 130 | + <div class="publication-links"> |
| 131 | + <span class="link-block"> |
| 132 | + <a href="https://arxiv.org/pdf/2011.12948" |
| 133 | + class="external-link button is-normal is-rounded is-dark"> |
| 134 | + <span class="icon"> |
| 135 | + <i class="fas fa-file-pdf"></i> |
| 136 | + </span> |
| 137 | + <span>Paper</span> |
| 138 | + </a> |
| 139 | + </span> |
| 140 | + <span class="link-block"> |
| 141 | + <a href="https://arxiv.org/abs/2011.12948" |
| 142 | + class="external-link button is-normal is-rounded is-dark"> |
| 143 | + <span class="icon"> |
| 144 | + <i class="ai ai-arxiv"></i> |
| 145 | + </span> |
| 146 | + <span>arXiv</span> |
| 147 | + </a> |
| 148 | + </span> |
| 149 | + <span class="link-block"> |
| 150 | + <a href="https://www.youtube.com/watch?v=MrKrnHhk8IA" |
| 151 | + class="external-link button is-normal is-rounded is-dark"> |
| 152 | + <span class="icon"> |
| 153 | + <i class="fab fa-youtube"></i> |
| 154 | + </span> |
| 155 | + <span>Video</span> |
| 156 | + </a> |
| 157 | + </span> |
| 158 | + <span class="link-block"> |
| 159 | + <a href="https://github.com/google/nerfies" |
| 160 | + class="external-link button is-normal is-rounded is-dark"> |
| 161 | + <span class="icon"> |
| 162 | + <i class="fab fa-github"></i> |
| 163 | + </span> |
| 164 | + <span>Code</span> |
| 165 | + </a> |
| 166 | + </span> |
| 167 | + <span class="link-block"> |
| 168 | + <a href="https://github.com/google/nerfies/releases/tag/0.1" |
| 169 | + class="external-link button is-normal is-rounded is-dark"> |
| 170 | + <span class="icon"> |
| 171 | + <i class="far fa-images"></i> |
| 172 | + </span> |
| 173 | + <span>Data</span> |
| 174 | + </a> |
| 175 | + </div> |
| 176 | + --> |
| 177 | + </div> |
| 178 | + </div> |
| 179 | + </div> |
| 180 | + </div> |
| 181 | + </div> |
| 182 | + </section> |
| 183 | + |
| 184 | + |
| 185 | + |
| 186 | + |
| 187 | + |
| 188 | + |
| 189 | + <section class="section"> |
| 190 | + <div class="container is-max-desktop"> |
| 191 | + <!-- Abstract. --> |
| 192 | + <div class="columns is-centered has-text-centered"> |
| 193 | + <div class="column is-four-fifths"> |
| 194 | + <h2 class="title is-3">Abstract</h2> |
| 195 | + <div class="content has-text-justified"> |
| 196 | + <p> |
| 197 | + Inverted indexes are the backbone of most large-scale information retrieval systems. |
| 198 | + Although conceptually simple, high-performance inverted indexes require a deep understanding |
| 199 | + of low-level system optimizations, compression techniques, and traversal strategies. With |
| 200 | + the widespread adoption of in-memory search engines, the rise of learned sparse retrieval |
| 201 | + (LSR), and the increasing complexity of ranking pipelines, the design space for efficient |
| 202 | + indexing and retrieval systems has expanded significantly. |
| 203 | + </p> |
| 204 | + <p> |
| 205 | + This tutorial addresses a critical knowledge gap between textbook-style explanations and |
| 206 | + advanced techniques required for efficient and optimized retrieval. It aims to equip |
| 207 | + researchers and practitioners with a comprehensive understanding of how modern in-memory |
| 208 | + search systems are designed, built, and optimized for high-performance retrieval across |
| 209 | + large-scale document collections. |
| 210 | + </p> |
| 211 | + |
| 212 | + </div> |
| 213 | + </div> |
| 214 | + </div> |
| 215 | + <!--/ Abstract. --> |
| 216 | + |
| 217 | + </section> |
| 218 | + |
| 219 | + <section class="section"> |
| 220 | + <div class="container is-max-desktop"> |
| 221 | + <div class="columns is-centered has-text-centered"> |
| 222 | + <div class="column is-four-fifths"> |
| 223 | + <h2 class="title is-3">Practical Component</h2> |
| 224 | + <div class="content has-text-justified"> |
| 225 | + <p> |
| 226 | + For the practical component, we will need to grab some data, and have some instructions ready. |
| 227 | + </p> |
| 228 | + <p> |
| 229 | + <strong>Prerequisite:</strong> You have your own machine with Docker installed. |
| 230 | + <br> |
| 231 | + I will step through on my own machine if you don’t have access. |
| 232 | + </p> |
| 233 | + <p> |
| 234 | + You can also experiment with the tutorial at any time (later). |
| 235 | + </p> |
| 236 | + <p> |
| 237 | + <strong>Tutorial:</strong> <a href="https://shorturl.at/VExpG" target="_blank">https://shorturl.at/VExpG</a> |
| 238 | + <br> |
| 239 | + <strong>AKA:</strong> <a href="https://github.com/pisa-engine/pisa/blob/main/test/docker/tutorial/instructions.md" target="_blank"> |
| 240 | + https://github.com/pisa-engine/pisa/blob/main/test/docker/tutorial/instructions.md</a> |
| 241 | + </p> |
| 242 | + <p> |
| 243 | + <strong>Slide Deck:</strong> <a href="https://pisa-engine.github.io/static/SIGIR2025-2.pdf" target="_blank">https://pisa-engine.github.io/static/SIGIR2025-2.pdf</a> |
| 244 | + </p> |
| 245 | + </div> |
| 246 | + </div> |
| 247 | + </div> |
| 248 | + </div> |
| 249 | + </section> |
| 250 | + |
| 251 | + <footer class="footer"> |
| 252 | + <div class="container"> |
| 253 | + <div class="content has-text-centered"> |
| 254 | + <a class="icon-link" href="./static/videos/nerfies_paper.pdf"> |
| 255 | + <i class="fas fa-file-pdf"></i> |
| 256 | + </a> |
| 257 | + <a class="icon-link" href="https://github.com/keunhong" class="external-link" disabled> |
| 258 | + <i class="fab fa-github"></i> |
| 259 | + </a> |
| 260 | + </div> |
| 261 | + <div class="columns is-centered"> |
| 262 | + <div class="column is-8"> |
| 263 | + <div class="content"> |
| 264 | + <p> |
| 265 | + This website is licensed under a <a rel="license" |
| 266 | + href="http://creativecommons.org/licenses/by-sa/4.0/">Creative |
| 267 | + Commons Attribution-ShareAlike 4.0 International License</a>. |
| 268 | + </p> |
| 269 | + <p> |
| 270 | + This means you are free to borrow the <a |
| 271 | + href="https://github.com/nerfies/nerfies.github.io">source code</a> of this website, |
| 272 | + we just ask that you link back to this page in the footer. |
| 273 | + Please remember to remove the analytics code included in the header of the website which |
| 274 | + you do not want on your website. |
| 275 | + </p> |
| 276 | + </div> |
| 277 | + </div> |
| 278 | + </div> |
| 279 | + </div> |
| 280 | + </footer> |
| 281 | + |
| 282 | +</body> |
| 283 | + |
| 284 | +</html> |
0 commit comments