Inference-Time Scaling for Joint Audio-Video Generation

arXiv:2606.0318388.5
AI Analysis

For researchers in multimodal generation, this work provides a training-free method to enhance joint audio-video generation quality, addressing the challenge of balancing multiple objectives.

This paper presents the first comprehensive study of inference-time scaling for joint audio-video generation, proposing a multi-verifier framework and an adaptive reward weighting algorithm that significantly improves semantic alignment, perceptual quality, and audio-visual synchronization on VGGSound and JavisBench-mini benchmarks.

Joint audio-video generation aims to synthesize realistic audio-video pairs that are both semantically aligned with text prompts and precisely synchronized. While existing joint audio-video generation models often require substantial training resources to improve fidelity, Inference-Time Scaling (ITS) has recently emerged as a promising training-free alternative in single-modality domains. However, extending ITS from a single modality to multimodal domains is non-trivial, as it requires balancing multiple heterogeneous objectives. In this paper, we present the first comprehensive study of ITS for joint audio-video generation. We first demonstrate that a multi-verifier framework is essential to address the limitations of single-objective guidance, including asymmetric performance trade-offs and verifier hacking. Through systematic analysis, we then identify an optimal multi-verifier combination that yields balanced improvements across all quality dimensions. Finally, to effectively aggregate diverse reward signals, we propose Adaptive Reward Weighting (ARW), a novel test-time optimization algorithm. ARW treats reward aggregation as an online optimization problem, utilizing learnable parameters to calibrate reward variances without requiring prior knowledge of reward distributions, thereby ensuring robust multi-objective selection. Experimental results on VGGSound and JavisBench-mini benchmarks demonstrate that our framework significantly enhances semantic alignment, perceptual quality, and audio-visual synchronization of generated outputs. Synthesized samples and code are available on the project page: https://jung-jaemin.github.io/ITS-AVGen-Proj.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes