AI Paper Atlas

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

$\alpha$-DPO: Robust Preference Alignment for Diffusion Models via $\alpha$ Divergence

Yang Li, Songlin Yang, Wei Wang, Xiaoxuan Han, Jing Dong

ICLR 2026 · Image generation › Text-to-image diffusion · approach: Preference optimization for diffusion · 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
replace forward-KL (DPO) with alpha-divergence objectiveSDXL · Pick-a-Pic test (trained on Pick-a-Pic V2)ImageReward: 0.9725 -> 1.054Improved
replace forward-KL (DPO) with alpha-divergence objectiveSD1.5 · Pick-a-Pic test (trained on Pick-a-Pic V2)HPSv2: 26.11 -> 27.59Improved
use alpha-DPO instead of DPO under 20% label flippingSDXL · Pick-a-Pic test (20% label flipping)ImageReward: 0.9102 -> 1.001Improved
use alpha-DPO instead of DPO under 20% label flippingSD1.5 · Pick-a-Pic test (20% label flipping)HPSv2: 25.86 -> 26.69Improved
add dynamic alpha scheduling (vs fixed alpha) (ablation)alpha-DPO SDXL · Pick-a-Pic testImageReward: 1.019 -> 1.054Improved
add dynamic alpha scheduling (vs fixed alpha) (ablation)alpha-DPO SDXL · Pick-a-Pic testPickScore: 22.45 -> 22.51Improved
raise mu from 0.1 to 0.9999 (ablation)alpha-DPO SDXL · Pick-a-Pic testImageReward: 1.013 -> 1.054Improved
raise mu from 0.1 to 0.9999 (ablation)alpha-DPO SDXL · Pick-a-Pic testPickScore: 22.31 -> 22.51Improved
start dynamic alpha at step 0 rather than step 200 (ablation)alpha-DPO SDXL · Pick-a-Pic testImageReward: 1.031 -> 1.054Improved
set beta to 4000 rather than 2000 (ablation)alpha-DPO SDXL · Pick-a-Pic testImageReward: 1.022 -> 1.054Improved
set beta to 4000 rather than 6000 (ablation)alpha-DPO SDXL · Pick-a-Pic testImageReward: 1.002 -> 1.054Improved
replace DPO component of SPO with alpha-DPOSPO SDXL · Pick-a-Pic testImageReward: 1.116 -> 1.134Improved

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