Inference-Time Alignment of Diffusion Models via Trust-Region Iterative Twisted Sequential Monte Carlo

ArXi:2605.25123v1 Announce Type: cross We study inference-time alignment for diffusion-based generative models, aiming to steer a base model toward high-reward outputs without updating its weights. Recent Sequential Monte Carlo (SMC)-based steering methods approximate reward-tilted target distributions in a principled way, but their proposals remain largely tied to the base sampler.