cs.CVMay 31, 2026

Knowledge-Intensive Video Generation

Authors: Chenxu Wang, Mingda Chen

Organizations: Fudan University · Shanghai Jiao Tong University

Abstract

Text-to-video generation has advanced rapidly in visual quality, but remains under-evaluated for factuality and practical usefulness. We introduce knowledge-intensive video generation (KIVI), where models generate videos from short information-seeking prompts that ask for explanations, procedures, or demonstrations. To evaluate this setting, we construct KIVI-Bench, a benchmark of 1,080 prompts, and propose automatic metrics for factuality and helpfulness. Human evaluation shows that our metrics significantly better align with human annotations than existing alternatives. Experiments on seven state-of-the-art video generation models show that current systems still lag behind human performance, especially on visual properties, procedural operations, and clear information presentation. These results highlight KIVI as a challenging direction for factual and instructionally useful video generation.

Explore similar work

CardsList
  1. VideoGen-Agent: Reinforcing Video Generation Agents

    Sep 21, 2026Binxu Li, Haoyi Duan, Yuhui Zhang +9Agentic Video Generation