arXiv Open Access 2026

Lookahead Sample Reward Guidance for Test-Time Scaling of Diffusion Models

Yeongmin Kim Donghyeok Shin Byeonghu Na Minsang Park Richard Lee Kim +1 lainnya
Lihat Sumber

Abstrak

Diffusion models have demonstrated strong generative performance; however, generated samples often fail to fully align with human intent. This paper studies a test-time scaling method that enables sampling from regions with higher human-aligned reward values. Existing gradient guidance methods approximate the expected future reward (EFR) at an intermediate particle $\mathbf{x}_t$ using a Taylor approximation, but this approximation at each time step incurs high computational cost due to sequential neural backpropagation. We show that the EFR at any $\mathbf{x}_t$ can be computed using only marginal samples from a pre-trained diffusion model. The proposed EFR formulation detaches the neural dependency between $\mathbf{x}_t$ and the EFR, enabling closed-form guidance computation without neural backpropagation. To further improve efficiency, we introduce lookahead sampling to collect marginal samples. For final sample generation, we use an accurate solver that guides particles toward high-reward lookahead samples. We refer to this sampling scheme as LiDAR sampling. LiDAR achieves substantial performance improvements using only three samples with a 3-step lookahead solver, exhibiting steep performance gains as lookahead accuracy and sample count increase; notably, it reaches the same GenEval performance as the latest gradient guidance method for SDXL with a 9.5x speedup.

Topik & Kata Kunci

Penulis (6)

Y

Yeongmin Kim

D

Donghyeok Shin

B

Byeonghu Na

M

Minsang Park

R

Richard Lee Kim

I

Il-Chul Moon

Format Sitasi

Kim, Y., Shin, D., Na, B., Park, M., Kim, R.L., Moon, I. (2026). Lookahead Sample Reward Guidance for Test-Time Scaling of Diffusion Models. https://arxiv.org/abs/2602.03211

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2026
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓