Rewind and Render: Towards Factually Accurate Text-to-Video Generation with Distilled Knowledge Retrieval

被引：0

作者：

Lee, Daniel ^{[1
]}

Chandra, Arjun ^{[2
]}

Zhou, Yang ^{[3
]}

Li, Yunyao ^{[1
]}

Conia, Simone ^{[4
]}

机构：

[1] Adobe, San Jose, CA 95110 USA

[2] Boston Univ, Boston, MA 02215 USA

[3] Adobe Res, San Francisco, CA USA

[4] Sapienza Univ Rome, Rome, Italy

来源：

THIRTY-NINTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, AAAI-25, VOL 39 NO 28 | 2025年

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Text-to-Video (T2V) models, despite recent advancements, struggle with factual accuracy, especially for knowledge-dense content. We introduce FACT-V (Factual Accuracy in Content Translation to Video), a system integrating multi-source knowledge retrieval into T2V pipelines. FACT-V offers two key benefits: i) improved factual accuracy of generated videos through dynamically retrieved information, and ii) increased interpretability by providing users with the augmented prompt information. A preliminary evaluation demonstrates the potential of knowledge-augmented approaches in improving the accuracy and reliability of T2V systems, particularly for entity-specific or time-sensitive prompts.

引用

页码：29652 / 29654

页数：3