AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

SD3.5-Flash: Distribution-Guided Distillation of Generative Flows

Hmrishav Bandyopadhyay, Rahim Entezari, Jim Scott, Reshinth Adithyan, Yi-Zhe Song, Varun Jampani

ECCV 2026 · Image generation › Fast diffusion sampling · approach: Rectified flow straightening · 16 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
share DMD timesteps with student trajectory timesteps (timestep sharing) (ablation)SD3.5-Flash 4-step · ablation (qualitative)image qualityImproved
add adversarial objective (ablation)SD3.5-Flash 4-step · ablation (qualitative)image qualityImproved
add trajectory-guidance pre-training of the student (ablation)SD3.5-Flash 4-step · ablation (qualitative)colour and compositionImproved
add periodic discriminator head refresh (ablation)SD3.5-Flash 4-step · ablation (qualitative)composition and smoothnessImproved
drop T5-XXL text encoder (null embeddings) (ablation)SD3.5-Flash 16-bit 4-step · COCO-30KGenEval: 0.70 -> 0.68Hurt
drop T5-XXL, 16-bit (ablation)SD3.5-Flash 4-step · peak VRAMpeak VRAM (GiB): 17.58 -> 8.71Improved
quantize MM-DiT 16-bit to 8-bit (no T5) (ablation)SD3.5-Flash 4-step · COCO-30KGenEval: 0.68 -> 0.68No clear effect
distill SD3.5M to 4 steps vs 50-step teacherSD3.5-Flash · COCO-30KGenEval: 0.64 -> 0.70Improved
distill SD3.5M to 4 steps vs 50-step teacherSD3.5-Flash · COCO-30KFID: 20.06 -> 29.80Hurt
split-timestep fine-tuning with 3:7 branch mergeSD3.5-Flash 4-step · GenEval prompt alignmentGenEvalImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy