/g/ — /sdg/ - Stable diffusion general

>mfw Resource news

06/10/2026

>EvoQuality: Self-Evolving VLM for Image Quality Assessment
https://huggingface.co/ByteDance/EvoQuality

>ComfyTV: Turn ComfyUI into a TapNow / LibTV-style canvas app
https://github.com/jtydhr88/ComfyTV

>PathRelax: Parallel-Path Relaxed Speculative Jacobi Decoding for Accelerating Auto-Regressive Text-to-Image Generation
https://github.com/Haodong-Lei-Ray/PathSpec

>SSR-Merge: Subspace Signal Routing for Training-Free LoRA Merging in Diffusion Models
https://github.com/nagara214/SSR-Merge

>SCAIL-2: Unifying Controlled Character Animation with End-to-end In-Context Conditioning
https://teal024.github.io/SCAIL-2

>IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder
https://github.com/Row11n/IDEAL

>Image to Prompt: Web app to turn an image into Ideogram 4 JSON prompt
https://github.com/cocktailpeanut/image-to-prompt

>Simple Diffusion XS (sdxs-2b alpha version)
https://huggingface.co/AiArtLab/sdxs-2b

>Bernini-R: Repackaged model files for ComfyUI
https://huggingface.co/Comfy-Org/Bernini-R

06/09/2026

>SCAIL-2: Unifying Controlled Character Animation with End-to-end In-Context Conditioning
https://teal024.github.io/SCAIL-2

>BLM-SGAN: Bidirectional Language Modeling for Semantic-Spatial Text-to-Image Generation
https://github.com/haidy-maher/BLM-SGAN-Text-to-Image-Generation

>SwiftVR: Real-Time One-Step Generative Video Restoration
https://h-oliday.github.io/SwiftVR

>Property-Informed Diffusion-Based Text-to-Microstructure Generation
https://github.com/hongsong-wang/PropDiff-TMG

>OmniTryOn: Video Try-On Anything at Once!
https://github.com/xcltql666/OminTryOn

>IEA: Amateur-Friendly Conversational Image Editing Agent via Three Stages of Multitask Alignment
https://github.com/OpenDFM/Image_Edit_Agent

>CapRL++: Unified Reinforcement Learning with Verifiable Rewards for Dense Image and Video Captioning
https://github.com/InternLM/CapRL

>mfw Research news

06/10/2026

>HarmoView: Harmonizing Multi-View Constraints for Identity-Consistent Video Generation
https://conallwang.github.io/HarmoView_Pages

>STEDiff: Strengthening Text Embedding for Text-to-Image Alignment in Diffusion Model
https://arxiv.org/abs/2606.10653

>FadeMem: Distance-Aware Memory Consolidation for Autoregressive Video Diffusion
https://arxiv.org/abs/2606.10671

>Making Time Editable in Video Diffusion Transformers
https://arxiv.org/abs/2606.10183

>BiWM: Advancing Open-Source Interactive Video World Models with Bidirectional Autoregression
https://arxiv.org/abs/2606.10135

>Mean Flow Distillation: Robust and Stable Distillation for Flow Matching Models
https://arxiv.org/abs/2606.11155

>Improving Text-Instance Alignment Of Foreground Conditioned Out-Painting Via Customized Concept Embedding
https://arxiv.org/abs/2606.10892

>FoA-SR: Faithful or Aesthetic? Profile-Aware Preference Optimization for Real-World Image Super-Resolution
https://arxiv.org/abs/2606.10275

>Dissect and Prune: Enhancing Robustness in AI-Generated Image Detection
https://arxiv.org/abs/2606.10309

>AnimaSpark: A Feed-Forward Method for Animating Arbitrary 3D Objects
https://arxiv.org/abs/2606.10988

>Lip Forcing: Few-Step Autoregressive Diffusion for Real-time Lip Synchronization
https://cvlab-kaist.github.io/LipForcing

>PTL-Diffusion: Manifold-Aware Diffusion with Periodic Terminal Laws
https://arxiv.org/abs/2606.09816

>Evaluating the Representation Space of Diffusion Models via Self-Supervised Principles
https://arxiv.org/abs/2606.09718

>Few-step Generative Models as Lossy Compression
https://arxiv.org/abs/2606.10450

>Pose-ICL: 3D-Aware In-Context Learning for Pose-Controllable Subject Customization
https://arxiv.org/abs/2606.10902

>Don't waste SAM
https://arxiv.org/abs/2606.10696

>containment general

>>109025445
This is why you are here every single day.

File: 1779676319471661.jpg (357 KB, 1338x784)

>>109025266 (OP)

File: debo_vn_fia_00076_.png (2.34 MB, 1792x977)

>>109025942
cool

>>109025945
even the raptors are shocked by the beheadings

File: 00000199-188977618490315-z-turbo-mt.jpg (1.45 MB, 3072x2048)

>>109025985
the beheadings will continue until morale improves

File: 00000007-1052011838727968-z-turbo-mt.jpg (1.42 MB, 3072x2048)

>>109026296
is she stuck?

File: 000000_74085_.png (2.28 MB, 1266x1292)

>>109026296
Nice.

File: 00043-comfyui-cfg=3.0-steps=7-ideogram4-2026-06-10-195635.jpg (438 KB, 1088x1936)

>>109026552
>>109026565
kek ty
she's fine

File: 000000_74103_.png (2.53 MB, 1280x1280)

>>109026645
kek nice wings.
>>109026660
check

File: 00000011-400291073912783-z-turbo-mt.jpg (1.40 MB, 3072x2048)

>>109026676
wut

File: 00061-comfyui-cfg=3.0-steps=7-ideogram4-2026-06-10-213358.jpg (601 KB, 1088x1936)

>>109026676
>kek nice wings.
ty. Capibaras deserve wings

gn

File: 00000017-893031659116311-z-turbo-mt.jpg (1.33 MB, 3072x2048)

>>109026725
gn

File: 00000032-516360651184681-z-turbo-mt.jpg (1.16 MB, 3072x2048)

gn all

File: debo_s_fia_00021_.png (1.64 MB, 1792x977)

>>109027257
gn

File: 000000_74153_.png (2.57 MB, 1054x1540)

>>109027829
Nice

G'mornin Anons, have a great day!

File: 00000038-841601240697423-z-turbo-mt.jpg (1006 KB, 3072x2048)

>>109028943
gm

i miss schizo anon

File: 00015-comfyui-cfg=3.0-steps=7-ideogram4-2026-06-11-085923.jpg (732 KB, 1456x1456)

I still have to figure things out in terms of object placement.

File: 00000047-122204910154112-z-turbo-mt.jpg (962 KB, 3072x2048)

>>109029574
lel
just use bbox bro

>>109029681
I'm trying to get the AI to generate the elements and correctly place the bbox but AIs aren't good at structured output.

File: 00000051-257530489314537-z-turbo-mt.jpg (1021 KB, 3072x2048)

>>109029800
maybe give it the resolution and tell it to place objects within those coordinates

File: 00024-comfyui-cfg=3.0-steps=7-ideogram4-2026-06-11-105305.jpg (371 KB, 1296x1616)

>>109029830
I think I got it - kek

File: 00000061-459375088508637-z-turbo-mt.jpg (1.45 MB, 3072x2048)

>>109030250
does ideogram have sameface?

File: debo_vn_fia_00077_.png (2.31 MB, 1792x977)

>>109029574
I like the surrealism of this a lot

File: 00028-comfyui-cfg=3.0-steps=7-ideogram4-2026-06-11-121757.jpg (457 KB, 1296x1616)

>>109030430
Dali was my inspiration.