3 papers
cs.LG2026
Aligning Few-Step Generative Models by Amortizing Sample-based Variational Inference
Jaewoo Lee, Hyeongyu Kang, Dohyun Kim +9
Aligning a few-step generative model is challenging, since existing alignment frameworks typically rely on restrictive assumptions: a tractable likelihood, a specific ODE/SDE solve…
cs.LG2026
Reward Score Matching: Unifying Reward-based Fine-tuning for Flow and Diffusion Models
Jeongjae Lee, Jinho Chang, Jeongsol Kim +1
Reward-based fine-tuning steers a pretrained diffusion or flow-based generative model toward higher-reward samples while remaining close to the pretrained model. Although existing…
cs.CV2025
PCPO: Proportionate Credit Policy Optimization for Aligning Image Generation Models
Jeongjae Lee, Jong Chul Ye
While reinforcement learning has advanced the alignment of text-to-image (T2I) models, state-of-the-art policy gradient methods are still hampered by training instability and high…