Pith. sign in

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

arxiv 2503.02039 v2 pith:BKELIVT2 submitted 2025-03-03 cs.LG cs.AI

classification cs.LGcs.AI
keywords diffusionalignmentdsearchinference-timerewardsearchacrossdomains
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
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.

Discussion (0). Sign in to comment.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Flash-BoN: Instant Drafts for Inference-Time Scaling in Diffusion Models

    cs.CV 2026-07 conditional novelty 6.0 of 10

    Under wall-clock budgets, cheap multi-knob drafts plus multi-stage verification outperform guided intermediate search for diffusion T2I inference-time scaling.

  2. Superbunched random fiber laser

    physics.optics 2026-03 unverdicted novelty 6.0 of 10

    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.

  3. Test-Time Alignment of Text-to-Image Diffusion Models via Null-Text Embedding Optimisation

    cs.CV 2025-11 conditional novelty 6.0 of 10

    Optimizing the null-text embedding in classifier-free guidance aligns diffusion outputs to a target reward while preserving cross-reward quality.

  4. Inference-Time Search Using Side Information for Diffusion-Based Image Reconstruction

    cs.CV 2025-10 conditional novelty 6.0 of 10

    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-...

  5. Diffusion Tree Sampling: Scalable inference-time alignment of diffusion models

    cs.LG 2025-06 conditional novelty 6.0 of 10

    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.

  6. Scaling Image and Video Generation via Test-Time Evolutionary Search

    cs.CV 2025-05 conditional novelty 6.0 of 10

    Evolutionary search over denoising trajectories improves image and video generation quality and diversity as test-time compute increases, without retraining the generative model.

  7. On the Design Space of Discrete Diffusion Online Adaptation for Molecular Optimization

    cs.LG 2026-07 conditional novelty 5.5 of 10

    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...

  8. Inference-Time Scaling of Diffusion Models via Progressive Seed Pruning

    cs.CV 2026-07 conditional novelty 5.0 of 10

    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...

  9. Ctrl-Z Sampling: Scaling Diffusion Sampling with Controlled Random Zigzag Explorations

    cs.CV 2025-06 conditional novelty 5.0 of 10

    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.

  10. Personalized Constitutionally-Aligned Agentic Superego: Secure AI Behavior Aligned to Diverse Human Values

    cs.AI 2025-06 conditional novelty 5.0 of 10

    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.

Pith tools