REVIEW 4 cited by
Test-time Alignment of Diffusion Models without Reward Over-optimization
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 excel in generative tasks, but aligning them with specific objectives while maintaining their versatility remains challenging. Existing fine-tuning methods often suffer from reward over-optimization, while approximate guidance approaches fail to optimize target rewards effectively. Addressing these limitations, we propose a training-free, test-time method based on Sequential Monte Carlo (SMC) to sample from the reward-aligned target distribution. Our approach, tailored for diffusion sampling and incorporating tempering techniques, achieves comparable or superior target rewards to fine-tuning methods while preserving diversity and cross-reward generalization. We demonstrate its effectiveness in single-reward optimization, multi-objective scenarios, and online black-box optimization. This work offers a robust solution for aligning diffusion models with diverse downstream objectives without compromising their general capabilities. Code is available at https://github.com/krafton-ai/DAS.
Forward citations
Cited by 4 Pith papers
-
Bootstrap Flow-Map Tree Sampling Enables Online Feedback Driven Search
Bootstrap Flow-Map Trees construct complete DDPM-like trajectories with a single NFE and dynamic steps, enabling efficient online feedback-driven search and alignment that beats prior tree and SMC samplers.
-
From Open Loop to Closed Loop: A Test-Time Iterative Optimization Framework for Reference-Consistent Image Generation
A training-free closed-loop PID controller iteratively corrects latent control signals so diffusion models stay consistent with ID, pose, or depth references better than matched open-loop sampling.
-
Post-Generation Curation of Synthetic Images via Homogeneous-Heterogeneous Splitting
Post-generation selection via Homogeneous-Heterogeneous real-data splits and a fidelity-diversity score raises synthetic-image utility for classification and segmentation without retraining generators.
-
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.
Discussion (0). Sign in to comment.