REVIEW 10 cited by
Dynamic Search for Inference-Time Alignment in Diffusion Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Diffusion models have shown promising generative capabilities across diverse domains, yet aligning their outputs with desired reward functions remains a challenge, particularly in cases where reward functions are non-differentiable. Some gradient-free guidance methods have been developed, but they often struggle to achieve optimal inference-time alignment. In this work, we newly frame inference-time alignment in diffusion as a search problem and propose Dynamic Search for Diffusion (DSearch), which subsamples from denoising processes and approximates intermediate node rewards. It also dynamically adjusts beam width and tree expansion to efficiently explore high-reward generations. To refine intermediate decisions, DSearch incorporates adaptive scheduling based on noise levels and a lookahead heuristic function. We validate DSearch across multiple domains, including biological sequence design, molecular optimization, and image generation, demonstrating superior reward optimization compared to existing approaches.
Forward citations
Cited by 10 Pith papers
-
Flash-BoN: Instant Drafts for Inference-Time Scaling in Diffusion Models
Under wall-clock budgets, cheap multi-knob drafts plus multi-stage verification outperform guided intermediate search for diffusion T2I inference-time scaling.
-
Superbunched random fiber laser
A fiber-integrated random laser uses Rayleigh scattering, cascaded Brillouin scattering, and four-wave mixing to generate multi-wavelength superbunched light with g(2)(0) up to ~26 and improved temporal ghost imaging.
-
Test-Time Alignment of Text-to-Image Diffusion Models via Null-Text Embedding Optimisation
Optimizing the null-text embedding in classifier-free guidance aligns diffusion outputs to a target reward while preserving cross-reward quality.
-
Inference-Time Search Using Side Information for Diffusion-Based Image Reconstruction
Injecting side information via inference-time particle search (GS/RFJS) improves diffusion-based inverse problem reconstructions across inpainting, super-resolution, deblurring, and MRI tasks in a training-free, plug-...
-
Diffusion Tree Sampling: Scalable inference-time alignment of diffusion models
Diffusion Tree Sampling is a Monte Carlo tree search over denoising trajectories that propagates terminal rewards backward to sample from reward-aligned distributions, showing up to 10x compute savings on tested benchmarks.
-
Scaling Image and Video Generation via Test-Time Evolutionary Search
Evolutionary search over denoising trajectories improves image and video generation quality and diversity as test-time compute increases, without retraining the generative model.
-
On the Design Space of Discrete Diffusion Online Adaptation for Molecular Optimization
Online fine-tuning of discrete diffusion models with complementary acquisition, CVaR shaping, density-entropy debiasing, replay, and validity control finds better molecules under fixed oracle budgets than offline fine...
-
Inference-Time Scaling of Diffusion Models via Progressive Seed Pruning
Progressive Seed Pruning—start many noise seeds, score early partially-denoised images, prune aggressively—improves prompt-aligned image generation at fixed denoising compute over best-of-N, resampling, and tree-searc...
-
Ctrl-Z Sampling: Scaling Diffusion Sampling with Controlled Random Zigzag Explorations
Ctrl-Z Sampling improves text-to-image outputs by adaptively rolling back and re-exploring when a reward model flags a quality plateau, at roughly 3 to 9 times the usual compute.
-
Personalized Constitutionally-Aligned Agentic Superego: Secure AI Behavior Aligned to Diverse Human Values
An external 'superego' module that filters agentic AI plans against user-selected 'constitutions' plus a universal safety floor is reported to cut harmful outputs by up to 98% on safety benchmarks.
Discussion (0). Sign in to comment.