AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Rethinking Direct Preference Optimization in Diffusion Models

Junyong Kang, Seohyun Lim, Kyungjune Baek, Hyunjung Shim

AAAI 2026 · Image generation › Text-to-image diffusion · approach: Preference optimization for diffusion · 2 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
add timestep-aware optimization on top of regularized reference-update exploration (ablation)Diffusion-DPO with reference update · Pick-a-Pic v2 test (SD1.5)PickScore: 21.88 -> 21.93Improved
add timestep-aware optimization alone to Diffusion-DPO (ablation)Diffusion-DPO · Pick-a-Pic v2 test (SD1.5)PickScore: 21.36 -> 21.10Hurt
add regularized reference-model update (exploration) to Diffusion-DPO (ablation)Diffusion-DPO · Pick-a-Pic v2 test (SD1.5)PickScore: 21.36 -> 21.88Improved
increase learning rate 1e-8 to 5e-8 to encourage exploration (ablation)Diffusion-DPO · Pick-a-Pic v2 test (SD1.5)PickScore: 21.36 -> 19.75Hurt
reduce beta from 5000 to 1000 (ablation)Diffusion-DPO · Pick-a-Pic v2 test (SD1.5)PickScore: 21.36 -> 21.26No clear effect
re-initialize reference to pretrained model instead of freezing at divergence boundary (ablation)reference update exploration · Pick-a-Pic v2 test (SD1.5)PickScore: 21.50 (Re-Init) vs 21.93 (Ours)Mixed
add reward scale schedule lambda(t) to timestep oversampling (gamma 0.9) (ablation)timestep-aware strategy · Pick-a-Pic v2 test (SD1.5)PickScore: 21.82 -> 21.93Improved
oversample early timesteps, gamma 0.9 vs 0.8 (ablation)timestep-aware strategy · Pick-a-Pic v2 test (SD1.5)ImageReward: 0.52 -> 0.64Improved
add divergence regularization to reference update across update periods tau 16/32/64 (ablation)reference update · Pick-a-Pic v2, PartiPrompts, HPDv2 (SD1.5)PickScoreImproved
full method vs strong baselines on SDXL, Aesthetic metric vs MaPOOurs · Pick-a-Pic v2 test (SDXL)Aesthetic win rate: 45.12Hurt

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