AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Contrastive Sampling Chains in Diffusion Models

Junyu Zhang, Daochang Liu, Shichao Zhang, Chang Xu

NeurIPS 2023 · Image generation › Fast diffusion sampling · approach: Diffusion theory and design space · 0 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
fine-tune with contrastive sampling chain (C++)EDM · CIFAR-10 unconditional, 35 NFEFID: 2.04 -> 1.88Improved
fine-tune with contrastive sampling chain (C++)EDM · CIFAR-10 unconditionalNFE to match FID: 35 -> 25 NFE at 2.04 FIDImproved
fine-tune with contrastive sampling chain (C++)LSGM · CIFAR-10 unconditional, 138 NFEFID: 2.10 -> 1.99Improved
fine-tune with contrastive sampling chain (C++)EDM · CIFAR-10 unconditional, 35 NFENLL: 2.60 -> 2.55Improved
fine-tune with contrastive sampling chain (C++)EDM · CIFAR-10 conditional, 35 NFEFID: 1.82 -> 1.73Improved
fine-tune with contrastive sampling chain (C++)EDM · FFHQ 64x64, 79 NFEFID: 2.39 -> 2.07Improved
fine-tune with contrastive sampling chain (C++)STDDPM · CelebA 64x64, 131 NFEFID: 1.90 -> 1.73Improved
fine-tune chain defined by DEIS sampler with C++DEIS-tAB3 on SDE VP · CIFAR-10, 10 NFEFID: 4.17 -> 4.02Improved
fine-tune chain defined by DPM-Solver-3 with C++DPM-Solver-3 on SDE VP · CIFAR-10, 10 NFEFID: 54.56 -> 43.81Improved
fine-tune chain defined by DEIS sampler with C++DEIS-tAB3 on IDDPM conditional · ImageNet 64x64, 10 NFEFID: 6.65 -> 6.59No clear effect
add BPTT to contrastive loss (linear schedule) (ablation)EDM contrastive fine-tuning · CIFAR-10FID: 1.98 -> 1.88Improved
compute contrastive loss over all chain steps instead of one pair (ablation)EDM contrastive fine-tuning · CIFAR-10, nonlinear scheduleFID: 1.99 -> 2.39Hurt

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