CutClaw: Agentic Hours-Long Video Editing via Music Synchronization

arXiv cs.CV / 4/1/2026

📰 NewsSignals & Early TrendsTools & Practical UsageModels & Research

共有:

Key Points

CutClaw is presented as an autonomous multi-agent framework that transforms hours of raw footage into short, meaningful videos with music synchronized editing.
The system uses hierarchical multimodal decomposition to capture both fine-grained visual details and global structure while also processing audio for alignment.
A “Playwriter Agent” coordinates narrative consistency over long horizons by anchoring visual scenes to musical shifts.
“Editor” and “Reviewer” agents collaborate to optimize the final cut using aesthetic and semantic criteria, improving the selection of fine-grained clips.
Experiments on hours-long-to-short generation report significant gains over state-of-the-art baselines, and the authors provide code via GitHub.

Abstract

Editing the video content with audio alignment forms a digital human-made art in current social media. However, the time-consuming and repetitive nature of manual video editing has long been a challenge for filmmakers and professional content creators alike. In this paper, we introduce CutClaw, an autonomous multi-agent framework designed to edit hours-long raw footage into meaningful short videos that leverages the capabilities of multiple Multimodal Language Models~(MLLMs) as an agent system. It produces videos with synchronized music, followed by instructions, and a visually appealing appearance. In detail, our approach begins by employing a hierarchical multimodal decomposition that captures both fine-grained details and global structures across visual and audio footage. Then, to ensure narrative consistency, a Playwriter Agent orchestrates the whole storytelling flow and structures the long-term narrative, anchoring visual scenes to musical shifts. Finally, to construct a short edited video, Editor and Reviewer Agents collaboratively optimize the final cut via selecting fine-grained visual content based on rigorous aesthetic and semantic criteria. We conduct detailed experiments to demonstrate that CutClaw significantly outperforms state-of-the-art baselines in generating high-quality, rhythm-aligned videos. The code is available at: https://github.com/GVCLab/CutClaw.