98 lines
No EOL
4 KiB
HTML
98 lines
No EOL
4 KiB
HTML
<!DOCTYPE html>
|
|
<html lang="en">
|
|
|
|
<head>
|
|
<meta charset="UTF-8">
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
|
<title>Yan Lin's Blog</title>
|
|
<link rel="icon" href="/logo.webp" type="image/x-icon">
|
|
<link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/css/bootstrap.min.css" rel="stylesheet">
|
|
<link href="https://cdn.jsdelivr.net/npm/bootstrap-icons@1.7.2/font/bootstrap-icons.css" rel="stylesheet">
|
|
<link rel="stylesheet" href="/index.css">
|
|
|
|
</head>
|
|
|
|
<body>
|
|
<main class="container">
|
|
|
|
<header class="border-bottom lh-1 py-3 border-secondary">
|
|
|
|
<div class="row flex-nowrap justify-content-between align-items-center">
|
|
<div class="col-2">
|
|
|
|
|
|
<a class="link-secondary header-icon px-2 h4" href="/"><i class="bi bi-house-fill"></i></a>
|
|
|
|
|
|
</div>
|
|
<div class="col-8 text-center">
|
|
<div class="page-header-logo h2 m-0 fw-bold" style="font-family: 'Abril Fatface', serif;">Yan Lin's Blog</div>
|
|
</div>
|
|
<div class="col-2 text-end">
|
|
|
|
|
|
|
|
</div>
|
|
</div>
|
|
|
|
|
|
</header>
|
|
|
|
|
|
<article class="section mt-4">
|
|
<div class="list-group list-group-flush">
|
|
|
|
<div class="list-group-item px-0">
|
|
<a class="mb-1 paper-title blog-link" href="/blog/html/one-step-diffusion-models.html">One Step Diffusion Models</a> | <span class="paper-title text-muted">May 2025</span>
|
|
<p class="card-text mb-auto tldr">Despite the promising performance of diffusion models on continuous modality generation, one deficiency that is holding them back is their requirement for multi-step denoising processes, which can be computationally expensive. In this article, we examine recent works that aim to build diffusion models capable of performing sampling in one or a few steps.</p>
|
|
</div>
|
|
|
|
<div class="list-group-item px-0">
|
|
<a class="mb-1 paper-title blog-link" href="/blog/html/multi-modal-transformer.html">Multi-modal and Multi-function Transformers</a> | <span class="paper-title text-muted">April 2025</span>
|
|
<p class="card-text mb-auto tldr">Multi-modal and multi-function Transformers enables a single architecture to process diverse data types such as language, images, and videos simultaneously. These models employ techniques like vector quantization and lookup-free quantization to map different modalities into a unified embedding space, allowing the Transformer to handle them within the same sequence. Beyond processing multiple data types, these architectures can also combine different functionalities-such as auto-regressive language generation and diffusion-based image creation-within a single model.</p>
|
|
</div>
|
|
|
|
</div>
|
|
</article>
|
|
|
|
</main>
|
|
|
|
<footer>
|
|
<div class="container">
|
|
<p class="text-center text-secondary" style="font-size: 0.8rem; font-family: 'Lato', sans-serif;">
|
|
Copyright © 2025. Designed and implemented by Yan Lin.
|
|
<span class="mx-1">|</span>
|
|
<a class="link link-secondary" target="_blank" href="https://git.yanlincs.com/yanlin/Homepage">Source Code</a>
|
|
</p>
|
|
</div>
|
|
</footer>
|
|
|
|
<button id="back-to-top" class="btn btn-light rounded-circle" style="position: fixed; bottom: 20px; right: 20px; display: none; z-index: 1000; width: 40px; height: 40px; padding: 0;"><i class="bi bi-chevron-up"></i></button>
|
|
|
|
|
|
<script>
|
|
// Show or hide the back-to-top button
|
|
window.addEventListener('scroll', function() {
|
|
var backToTopButton = document.getElementById('back-to-top');
|
|
if (window.scrollY > 100) {
|
|
backToTopButton.style.display = 'block';
|
|
} else {
|
|
backToTopButton.style.display = 'none';
|
|
}
|
|
});
|
|
|
|
// Scroll to top when the button is clicked
|
|
document.getElementById('back-to-top').addEventListener('click', function(e) {
|
|
e.preventDefault();
|
|
window.scrollTo({
|
|
top: 0,
|
|
behavior: 'smooth'
|
|
});
|
|
window.location.href = '#';
|
|
return false;
|
|
});
|
|
</script>
|
|
|
|
</body>
|
|
|
|
</html> |