PrismAudio: Decomposed Chain-of-Thoughts and Multi-dimensional Rewards for Video-to-Audio Generation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Liu, Huadai, Luo, Kaicheng, Wang, Wen, Chen, Qian, Sun, Peiwen, Huang, Rongjie, Li, Xiangang, Ye, Jieping, Xue, Wei
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911480788549632
author Liu, Huadai
Luo, Kaicheng
Wang, Wen
Chen, Qian
Sun, Peiwen
Huang, Rongjie
Li, Xiangang
Ye, Jieping
Xue, Wei
author_facet Liu, Huadai
Luo, Kaicheng
Wang, Wen
Chen, Qian
Sun, Peiwen
Huang, Rongjie
Li, Xiangang
Ye, Jieping
Xue, Wei
contents Video-to-Audio (V2A) generation requires balancing four critical perceptual dimensions: semantic consistency, audio-visual temporal synchrony, aesthetic quality, and spatial accuracy; yet existing methods suffer from objective entanglement that conflates competing goals in single loss functions and lack human preference alignment. We introduce PrismAudio, the first framework to integrate Reinforcement Learning into V2A generation with specialized Chain-of-Thought (CoT) planning. Our approach decomposes monolithic reasoning into four specialized CoT modules (Semantic, Temporal, Aesthetic, and Spatial CoT), each paired with targeted reward functions. This CoT-reward correspondence enables multidimensional RL optimization that guides the model to jointly generate better reasoning across all perspectives, solving the objective entanglement problem while preserving interpretability. To make this optimization computationally practical, we propose Fast-GRPO, which employs hybrid ODE-SDE sampling that dramatically reduces the training overhead compared to existing GRPO implementations. We also introduce AudioCanvas, a rigorous benchmark that is more distributionally balanced and covers more realistically diverse and challenging scenarios than existing datasets, with 300 single-event classes and 501 multi-event samples. Experimental results demonstrate that PrismAudio achieves state-of-the-art performance across all four perceptual dimensions on both the in-domain VGGSound test set and out-of-domain AudioCanvas benchmark. The project page is available at https://PrismAudio.github.io.
format Preprint
id arxiv_https___arxiv_org_abs_2511_18833
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PrismAudio: Decomposed Chain-of-Thoughts and Multi-dimensional Rewards for Video-to-Audio Generation
Liu, Huadai
Luo, Kaicheng
Wang, Wen
Chen, Qian
Sun, Peiwen
Huang, Rongjie
Li, Xiangang
Ye, Jieping
Xue, Wei
Sound
Computer Vision and Pattern Recognition
Audio and Speech Processing
Image and Video Processing
Video-to-Audio (V2A) generation requires balancing four critical perceptual dimensions: semantic consistency, audio-visual temporal synchrony, aesthetic quality, and spatial accuracy; yet existing methods suffer from objective entanglement that conflates competing goals in single loss functions and lack human preference alignment. We introduce PrismAudio, the first framework to integrate Reinforcement Learning into V2A generation with specialized Chain-of-Thought (CoT) planning. Our approach decomposes monolithic reasoning into four specialized CoT modules (Semantic, Temporal, Aesthetic, and Spatial CoT), each paired with targeted reward functions. This CoT-reward correspondence enables multidimensional RL optimization that guides the model to jointly generate better reasoning across all perspectives, solving the objective entanglement problem while preserving interpretability. To make this optimization computationally practical, we propose Fast-GRPO, which employs hybrid ODE-SDE sampling that dramatically reduces the training overhead compared to existing GRPO implementations. We also introduce AudioCanvas, a rigorous benchmark that is more distributionally balanced and covers more realistically diverse and challenging scenarios than existing datasets, with 300 single-event classes and 501 multi-event samples. Experimental results demonstrate that PrismAudio achieves state-of-the-art performance across all four perceptual dimensions on both the in-domain VGGSound test set and out-of-domain AudioCanvas benchmark. The project page is available at https://PrismAudio.github.io.
title PrismAudio: Decomposed Chain-of-Thoughts and Multi-dimensional Rewards for Video-to-Audio Generation
topic Sound
Computer Vision and Pattern Recognition
Audio and Speech Processing
Image and Video Processing
url https://arxiv.org/abs/2511.18833