[2410.14324v1] HiCo: Hierarchical Controllable Diffusion Model for Layout-to-image Generation
https://arxiv.org/abs/2410.14324v1
[2412.20292] An analytic theory of creativity in convolutional diffusion models
https://arxiv.org/abs/2412.20292
“Teaching language models to handle images without messing up their text abilities LMFusion enables text-only LLMs to understand and generate both text and images while preserving their original language capabilities through modality-specific processing. —– 🤔 Original
https://x.com/rohanpaul_ai/status/1876213048261874079
“🎨Text-to-Image Arena Leaderboard is now live with 40K+ community votes! Top Models: – #1. Recraft V3 – #2. Ideogram 2.0 – #3. FLUX1.1 [pro] – #3. Luma Photon – #5. DALL·E 3 – #5. FLUX.1 [dev] – #7. Stable Diffusion 3.5 Large Congrats to @recraftai @ideogram_ai @bfl_ml
https://x.com/lmarena_ai/status/1876318670621901018
“A collection of demos and example applications for Transformers.js, including text embeddings, sentiment analysis, image segmentation, and more, in various JavaScript environments like Node.js, Deno, and WebGPU
https://x.com/tom_doerr/status/1877343672280207668
“my benchmark might be broken but libvips seems to be 25x faster at resizing images compared to pillow” / X
https://x.com/vikhyatk/status/1875200315966005513
“Edit any image freely with detailed control, From object removal to adding new elements—BrushEdit merges MLLMs and inpainting for seamless, user-defined image edits. Paper: “BrushEdit: All-In-One Image Inpainting and Editing” → Addresses the limitations of inversion-based
https://x.com/rohanpaul_ai/status/1876950083742183787
