https://hackerpaper.kaipfister.de/p/2409.11340/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=41999115urn:hackerpapers:mention:2409.11340:41999115Wed, 30 Oct 2024 19:21:25 GMT<p>Elegant architecture, trained from scratch, excels at image editing. This looks very interesting! From https://arxiv.org/html/2409.11340v1 &gt; Unlike popular diffusion models, OmniGen features a very concise structure, comprising only two main components: a VAE and a transformer model, without any additional encoders. &gt; OmniGen supports arbitrarily interleaved text and image inputs as conditions to guide image generation, rather than text-only or image-only conditions. &gt; Additionally, we incorporate several classic computer vision tasks such as human pose estimation, edge detection, and image…</p><p><a href="https://hackerpaper.kaipfister.de/p/2409.11340/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CV