Sparse-LaViDa: Sparse Multimodal Discrete Diffusion Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Shufan, Gu, Jiuxiang, Liu, Kangning, Lin, Zhe, Wei, Zijun, Grover, Aditya, Kuen, Jason
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908714474143744
author Li, Shufan
Gu, Jiuxiang
Liu, Kangning
Lin, Zhe
Wei, Zijun
Grover, Aditya
Kuen, Jason
author_facet Li, Shufan
Gu, Jiuxiang
Liu, Kangning
Lin, Zhe
Wei, Zijun
Grover, Aditya
Kuen, Jason
contents Masked Discrete Diffusion Models (MDMs) have achieved strong performance across a wide range of multimodal tasks, including image understanding, generation, and editing. However, their inference speed remains suboptimal due to the need to repeatedly process redundant masked tokens at every sampling step. In this work, we propose Sparse-LaViDa, a novel modeling framework that dynamically truncates unnecessary masked tokens at each inference step to accelerate MDM sampling. To preserve generation quality, we introduce specialized register tokens that serve as compact representations for the truncated tokens. Furthermore, to ensure consistency between training and inference, we design a specialized attention mask that faithfully matches the truncated sampling procedure during training. Built upon the state-of-the-art unified MDM LaViDa-O, Sparse-LaViDa achieves up to a 2x speedup across diverse tasks including text-to-image generation, image editing, and mathematical reasoning, while maintaining generation quality.
format Preprint
id arxiv_https___arxiv_org_abs_2512_14008
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Sparse-LaViDa: Sparse Multimodal Discrete Diffusion Language Models
Li, Shufan
Gu, Jiuxiang
Liu, Kangning
Lin, Zhe
Wei, Zijun
Grover, Aditya
Kuen, Jason
Computer Vision and Pattern Recognition
Masked Discrete Diffusion Models (MDMs) have achieved strong performance across a wide range of multimodal tasks, including image understanding, generation, and editing. However, their inference speed remains suboptimal due to the need to repeatedly process redundant masked tokens at every sampling step. In this work, we propose Sparse-LaViDa, a novel modeling framework that dynamically truncates unnecessary masked tokens at each inference step to accelerate MDM sampling. To preserve generation quality, we introduce specialized register tokens that serve as compact representations for the truncated tokens. Furthermore, to ensure consistency between training and inference, we design a specialized attention mask that faithfully matches the truncated sampling procedure during training. Built upon the state-of-the-art unified MDM LaViDa-O, Sparse-LaViDa achieves up to a 2x speedup across diverse tasks including text-to-image generation, image editing, and mathematical reasoning, while maintaining generation quality.
title Sparse-LaViDa: Sparse Multimodal Discrete Diffusion Language Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.14008