Diverse Video Generation with Determinantal Point Process-Guided Policy Optimization

TL;DR

This paper introduces DPP-GRPO, a novel framework combining Determinantal Point Processes and Group Relative Policy Optimization to enhance diversity in text-to-video generation without compromising quality.

Contribution

It proposes a model-agnostic, set-level policy optimization approach that explicitly encourages diverse video outputs for a given prompt, addressing low-diversity issues in current T2V models.

Findings

Improves video diversity on state-of-the-art benchmarks

Maintains prompt fidelity and perceptual quality

Provides a new dataset of 30,000 diverse prompts

Abstract

While recent text-to-video (T2V) diffusion models have achieved impressive quality and prompt alignment, they often produce low-diversity outputs when sampling multiple videos from a single text prompt. We tackle this challenge by formulating it as a set-level policy optimization problem, with the goal of training a policy that can cover the diverse range of plausible outcomes for a given prompt. To address this, we introduce DPP-GRPO, a novel framework for diverse video generation that combines Determinantal Point Processes (DPPs) and Group Relative Policy Optimization (GRPO) theories to enforce explicit reward on diverse generations. Our objective turns diversity into an explicit signal by imposing diminishing returns on redundant samples (via DPP) while supplies groupwise feedback over candidate sets (via GRPO). Our framework is plug-and-play and model-agnostic, and encourages diverse generations across visual appearance, camera motions, and scene structure without sacrificing prompt fidelity or perceptual quality. We implement our method on WAN and CogVideoX, and show that our method consistently improves video diversity on state-of-the-art benchmarks such as VBench, VideoScore, and human preference studies. Moreover, we release our code and a new benchmark dataset of 30,000 diverse prompts to support future research.