---
format: "aidr-story-markdown/v1"
id: "e019fea75c08c1c6578c2410dfb4921d44a8c0cbbf5dbdd0916dfdd61a4fb7f5"
canonical_url: "https://aidr.today/e019fea7?lang=en"
title: "WeVisCut: The Whole Editing Room in One Agent From Narration to Long-Form Video"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-30T00:00:00.000Z"
category: "Research"
topics: ["agent"]
source_urls: ["https://huggingface.co/papers/2610.38352"]
summary: "WeVisCut is a long-form script-to-video system that connects tool execution with policy learning, using a four-stage data pipeline applied to 3,750 finished videos to generate over 130K training examples for a 9B policy model."
---

# WeVisCut: The Whole Editing Room in One Agent From Narration to Long\-Form Video

> [Open the canonical story](<https://aidr.today/e019fea7?lang=en>)

**Published:** 2026-10-30T00:00:00.000Z
**Category:** Research
**Topics:** agent

## Summary

WeVisCut is a long\-form script\-to\-video system that connects tool execution with policy learning, using a four\-stage data pipeline applied to 3,750 finished videos to generate over 130K training examples for a 9B policy model\.

## Sources

- [Story source](<https://huggingface.co/papers/2610.38352>)