Verifier Threshold: An Efficient Test-Time Scaling Approach for Image Generation
This addresses efficiency issues for users of large image generation models, though it is an incremental improvement over existing methods.
The paper tackles the problem of inefficient test-time compute allocation in image generation models by proposing Verifier-Threshold, which automatically reallocates compute to reduce computational time by 2-4x while maintaining performance on the GenEval benchmark.
Image generation has emerged as a mainstream application of large generative models. Just as test-time compute and reasoning have improved language model capabilities, similar benefits have been observed for image generation models. In particular, searching over noise samples for diffusion and flow models has been shown to scale well with test-time compute. While recent works explore allocating non-uniform inference-compute budgets across denoising steps, existing approaches rely on greedy heuristics and often allocate the compute budget ineffectively. In this work, we study this problem and propose a simple fix. We propose Verifier-Threshold, which automatically reallocates test-time compute and delivers substantial efficiency improvements. For the same performance on the GenEval benchmark, we achieve a 2-4x reduction in computational time over the state-of-the-art method.