---
format: "aidr-story-markdown/v1"
id: "3d4523482971df05fa6c52d889cbecf859069973366311f7d09efd2e0f5b03a1"
canonical_url: "https://aidr.today/3d452348?lang=en"
title: "JoVox: Joint Hierarchical Flow Matching for Lightweight Voxel Generation"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-10T00:00:00.000Z"
category: "Research"
topics: ["research"]
source_urls: ["https://huggingface.co/papers/2611.10690"]
summary: "JoVox is a discrete flow matching framework that jointly learns the distribution of all levels in a voxel hierarchy to improve lightweight voxel generation. By allowing finer cells to influence coarse cell prediction and using two-level prediction, it reduces the number of levels processed and increases voxel fidelity."
---

# JoVox: Joint Hierarchical Flow Matching for Lightweight Voxel Generation

> [Open the canonical story](<https://aidr.today/3d452348?lang=en>)

**Published:** 2026-11-10T00:00:00.000Z
**Category:** Research
**Topics:** research

## Summary

JoVox is a discrete flow matching framework that jointly learns the distribution of all levels in a voxel hierarchy to improve lightweight voxel generation\. By allowing finer cells to influence coarse cell prediction and using two\-level prediction, it reduces the number of levels processed and increases voxel fidelity\.

## Sources

- [Story source](<https://huggingface.co/papers/2611.10690>)