Motif-Video 2B: Technical Report

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lim, Junghwan, Cheung, Wai Ting, Ha, Minsu, Kim, Beomgyu, Kim, Taewhan, Lee, Haesol, Oh, Dongpin, Lee, Jeesoo, Kim, Taehyun, Kim, Minjae, Lee, Sungmin, Cho, Hyeyeon, Choi, Dahye, Her, Jaeheui, Huh, Jaeyeon, Jung, Hanbin, Kang, Changjin, Kim, Dongseok, Kim, Jangwoong, Kim, Youngrok, Kweon, Hyukjin, Lee, Hongjoo, Lee, Jeongdoo, Lee, Junhyeok, Park, Eunhwan, Park, Yeongjae, Ryu, Bokki, Weon, Dongjoo
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916023943299072
author Lim, Junghwan
Cheung, Wai Ting
Ha, Minsu
Kim, Beomgyu
Kim, Taewhan
Lee, Haesol
Oh, Dongpin
Lee, Jeesoo
Kim, Taehyun
Kim, Minjae
Lee, Sungmin
Cho, Hyeyeon
Choi, Dahye
Her, Jaeheui
Huh, Jaeyeon
Jung, Hanbin
Kang, Changjin
Kim, Dongseok
Kim, Jangwoong
Kim, Youngrok
Kweon, Hyukjin
Lee, Hongjoo
Lee, Jeongdoo
Lee, Junhyeok
Park, Eunhwan
Park, Yeongjae
Ryu, Bokki
Weon, Dongjoo
author_facet Lim, Junghwan
Cheung, Wai Ting
Ha, Minsu
Kim, Beomgyu
Kim, Taewhan
Lee, Haesol
Oh, Dongpin
Lee, Jeesoo
Kim, Taehyun
Kim, Minjae
Lee, Sungmin
Cho, Hyeyeon
Choi, Dahye
Her, Jaeheui
Huh, Jaeyeon
Jung, Hanbin
Kang, Changjin
Kim, Dongseok
Kim, Jangwoong
Kim, Youngrok
Kweon, Hyukjin
Lee, Hongjoo
Lee, Jeongdoo
Lee, Junhyeok
Park, Eunhwan
Park, Yeongjae
Ryu, Bokki
Weon, Dongjoo
contents Training strong video generation models usually requires massive datasets, large parameter counts, and substantial compute. In this work, we ask whether strong text-to-video quality is possible at a much smaller budget: fewer than 10M clips and less than 100,000 H200 GPU hours. Our core claim is that part of the answer lies in how model capacity is organized, not only in how much of it is used. In video generation, prompt alignment, temporal consistency, and fine-detail recovery can interfere with one another when they are handled through the same pathway. Motif-Video 2B addresses this by separating these roles architecturally, rather than relying on scale alone. The model combines two key ideas. First, Shared Cross-Attention strengthens text control when video token sequences become long. Second, a three-part backbone separates early fusion, joint representation learning, and detail refinement. To make this design effective under a limited compute budget, we pair it with an efficient training recipe based on dynamic token routing and early-phase feature alignment to a frozen pretrained video encoder. Our analysis shows that later blocks develop clearer cross-frame attention structure than standard single-stream baselines. On VBench, Motif-Video~2B reaches 83.76\%, surpassing Wan2.1 14B while using 7$\times$ fewer parameters and substantially less training data. These results suggest that careful architectural specialization, combined with an efficiency-oriented training recipe, can narrow or exceed the quality gap typically associated with much larger video models.
format Preprint
id arxiv_https___arxiv_org_abs_2604_16503
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Motif-Video 2B: Technical Report
Lim, Junghwan
Cheung, Wai Ting
Ha, Minsu
Kim, Beomgyu
Kim, Taewhan
Lee, Haesol
Oh, Dongpin
Lee, Jeesoo
Kim, Taehyun
Kim, Minjae
Lee, Sungmin
Cho, Hyeyeon
Choi, Dahye
Her, Jaeheui
Huh, Jaeyeon
Jung, Hanbin
Kang, Changjin
Kim, Dongseok
Kim, Jangwoong
Kim, Youngrok
Kweon, Hyukjin
Lee, Hongjoo
Lee, Jeongdoo
Lee, Junhyeok
Park, Eunhwan
Park, Yeongjae
Ryu, Bokki
Weon, Dongjoo
Computer Vision and Pattern Recognition
Artificial Intelligence
Training strong video generation models usually requires massive datasets, large parameter counts, and substantial compute. In this work, we ask whether strong text-to-video quality is possible at a much smaller budget: fewer than 10M clips and less than 100,000 H200 GPU hours. Our core claim is that part of the answer lies in how model capacity is organized, not only in how much of it is used. In video generation, prompt alignment, temporal consistency, and fine-detail recovery can interfere with one another when they are handled through the same pathway. Motif-Video 2B addresses this by separating these roles architecturally, rather than relying on scale alone. The model combines two key ideas. First, Shared Cross-Attention strengthens text control when video token sequences become long. Second, a three-part backbone separates early fusion, joint representation learning, and detail refinement. To make this design effective under a limited compute budget, we pair it with an efficient training recipe based on dynamic token routing and early-phase feature alignment to a frozen pretrained video encoder. Our analysis shows that later blocks develop clearer cross-frame attention structure than standard single-stream baselines. On VBench, Motif-Video~2B reaches 83.76\%, surpassing Wan2.1 14B while using 7$\times$ fewer parameters and substantially less training data. These results suggest that careful architectural specialization, combined with an efficiency-oriented training recipe, can narrow or exceed the quality gap typically associated with much larger video models.
title Motif-Video 2B: Technical Report
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2604.16503