arXiv Machine Learning October 16, 2026 By Woocheol Shin, Taeyoung Yun, Jiwon Sung, Hyeongyu Kang, Jeongmin Park, Jaewoo Lee, Kyuil Sim, Dohyun Kim, Yutong He, Jinkyoo Park

Noise-Space Reward Alignment for Few-Step Generative Models

Read the original on arXiv Machine Learning →

The paper introduces Noise Policy Optimization (NPO), a method for aligning few-step generative models with black-box reward signals. NPO learns an amortized flow policy that transforms initial Gaussian noise into a reward-tilted distribution, allowing the pretrained generator to produce higher-reward outputs without fine-tuning. The approach achieves superior reward-quality trade-offs across image, language, and protein generation tasks while drastically reducing inference cost compared to test-time search methods.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv AI

Jul 7

Trust-Region Noise Search for Black-Box Alignment of Diffusion and Flow Models

arXiv:2603. 14504v2 Announce Type: replace-cross Abstract: Optimizing the noise samples of diffusion and flow models is an increasingly popular approach to align these models to target rewards at inference time.

By Niklas Schweiger, Daniel Cremers, Karnik Ram

diffusion safety

arXiv AI

Oct 2

Manifold-Constrained Initial Noise Optimization for Efficient Generative Model Alignment

The paper introduces ZeNOVA, a gradient‑free method for aligning initial noise in generative models. It uses annealed soft‑value guidance, manifold‑constrained hyperspherical Langevin dynamics, and Metropolis‑Hastings jumps to address instability in black‑box reward settings. Experiments on image and video models show ZeNOVA outperforms existing zeroth‑order baselines by more stably optimizing noise toward higher rewards.

By Jinho Chang, Jong Chul Ye

efficiency safety

arXiv Machine Learning

Jun 19

Meta Flow Maps enable scalable reward alignment

arXiv:2601. 14430v2 Announce Type: replace-cross Abstract: Controlling generative models is computationally expensive.

By Peter Potaptchik, Adhi Saravanan, Abbas Mammadov, Alvaro Prat, Michael S. Albergo, Yee Whye Teh

fine-tuning safety

Hugging Face Trending Papers

Sep 30

Manifold-Constrained Initial Noise Optimization for Efficient Generative Model Alignment

Recent advances in distillation and flow-map models have enabled deterministic one- or few-step generation for high-quality data, facilitating a new branch of reward alignment approaches that directly...

efficiency safety

arXiv Machine Learning

Aug 28

GRAS: Guided Reduced-Variance Proposals and Adaptive Selection for Training-Free Reward Alignment in Discrete Diffusion

The paper introduces GRAS, a method that improves training‑free reward alignment for discrete diffusion models by reducing variance in guided proposals and adapting the resampling temperature during search. It achieves this without adding denoiser cost, using Rao‑Blackwellized estimates for differentiable rewards and a leave‑one‑out baseline for non‑differentiable ones. Experiments on regulatory DNA and protein design show GRAS outperforms existing training‑free techniques and rivals reward‑fine‑tuned models.

By Kwanyoung Kim

diffusion fine-tuning safety

arXiv Machine Learning

Jun 10

Gradient-Guided Reward Optimization for Inference-time Alignment

arXiv:2606. 09635v1 Announce Type: cross Abstract: Ensuring the reliability of Large Language Models (LLMs) under distribution drift requires inference-time adaptation.

By Hankun Lin, Ruqi Zhang

llms reinforcement-learning benchmarks safety

One email a morning, machine-written

Related stories

Trust-Region Noise Search for Black-Box Alignment of Diffusion and Flow Models

Manifold-Constrained Initial Noise Optimization for Efficient Generative Model Alignment

Meta Flow Maps enable scalable reward alignment

Manifold-Constrained Initial Noise Optimization for Efficient Generative Model Alignment

GRAS: Guided Reduced-Variance Proposals and Adaptive Selection for Training-Free Reward Alignment in Discrete Diffusion

Gradient-Guided Reward Optimization for Inference-time Alignment