PhD Student @ CompVis group, LMU Munich Working on diffusion & flow models🫶https://bsky.app/profile/joh-schb.bsky.socialhttps://bsky.app/profile/joh-schb.bsky.social/post/3m3f2fj4lbk2h🤔 What if you could generate an entire image using just one continuous token? 💡 It works if we leverage a self-supervised representation! Meet RepTok🦎: A generative model that encodes an image into a single continuous latent while keeping realism and semantics. 🧵 👇17 Oct 2025 10:20 +0000at://did:plc:a7vrncdbln2ofluqjomdlmlt/app.bsky.feed.post/3m3f2fj4lbk2hhttps://bsky.app/profile/joh-schb.bsky.social/post/3lqx6npb6h225Looking forward to attending #CVPR2025 in Nashville next week 🎸🎶 @mgui7.bsky.social and I will be presenting our latest work: 🌊 Diff2Flow: Training Flow Matching Models via Diffusion Model Alignment06 Jun 2025 15:47 +0000at://did:plc:a7vrncdbln2ofluqjomdlmlt/app.bsky.feed.post/3lqx6npb6h225https://bsky.app/profile/joh-schb.bsky.social/post/3ljtt2jwcic23Sunrise in the office after the #ICCV deadline night with @mgui7.bsky.social 🚀08 Mar 2025 05:46 +0000at://did:plc:a7vrncdbln2ofluqjomdlmlt/app.bsky.feed.post/3ljtt2jwcic23https://bsky.app/profile/joh-schb.bsky.social/post/3lcno4czb5c24Congrats to @frankfundel.bsky.social for publishing this work at WACV🔥 Has been a pleasure to jointly work on this topic with such a talented master student🤗 Looking forward to seeing what comes next!🚀 [contains quote post or other embedded content]06 Dec 2024 17:05 +0000at://did:plc:a7vrncdbln2ofluqjomdlmlt/app.bsky.feed.post/3lcno4czb5c24https://bsky.app/profile/joh-schb.bsky.social/post/3lck47pfx4s2eAwesome work from some colleagues cleaning up diffusion features!🚀 [contains quote post or other embedded content]05 Dec 2024 07:06 +0000at://did:plc:a7vrncdbln2ofluqjomdlmlt/app.bsky.feed.post/3lck47pfx4s2e