VideoGen Model 1: Prompt Expansion Versus Formatting in Audiovisual Generation

David Grossman; Anton Koenig

cs.CVOct 21, 2026

VideoGen Model 1: Prompt Expansion Versus Formatting in Audiovisual Generation

Authors: David Grossman, Anton Koenig

Organizations: VideoGen

Abstract

Creative prompt expansion improved overall audiovisual quality scores over raw prompts in a controlled pilot of VideoGen Model 1, our post-trained version of MiniMax H3. We compared the raw, reformatted, and expanded versions of 30 prompts, holding the generator, seed, and sampling settings fixed. Each clip was rated separately on five dimensions. Their mean formed an overall score on a 1-10 scale. Compared with formatting alone, overall scores increased by 0.63 points (95% interval: [+0.11, +1.15]) under VA-Judger and 0.15 points (95% interval: [-0.03, +0.35]) under Gemini 3.8 Flash, whose overall change remained uncertain. Prompt-alignment scores increased by 0.97 points under VA-Judger and 0.50 under Gemini. VA-Judger also showed higher audio quality; visual-quality ratings changed little under either judge.

Figures & tables

Dimension	VA-Judger	Gemini 3.8 Flash
Prompt alignment	$+0.97\;[+0.23,+1.73]$	$+0.50\;[+0.13,+0.93]$
Audiovisual consistency	$+0.50\;[-0.07,+1.10]$	$+0.10\;[-0.13,+0.37]$
Audio quality	$+1.00\;[+0.27,+1.80]$	$+0.17\;[-0.07,+0.43]$
Video quality	$+0.07\;[-0.43,+0.53]$	$-0.03\;[-0.27,+0.20]$
Completeness/coherence	$+0.60\;[0.00,+1.23]$	$0.00\;[-0.27,+0.33]$

Table 1: Paired mean differences between expanded and format-only prompts on each 1–10 dimension. Positive values favor expansion. Brackets give descriptive 95% paired-bootstrap intervals. Every dimension uses all 30 prompts.

Figure 2: Paired mean changes in overall score across all 30 prompts. Positive changes favor the first condition named. Whiskers show descriptive 95% paired-bootstrap intervals. The dashed line marks zero change. Ratings are not calibrated between judges.

Text-to-Audio-Video Generation Evaluation

See all 8 papers

ProactGen: Learning Proactive Clarification for Video Generation
Oct 15, 2026Zhixiao Wang, Yang Chen, Yifan Xu +5Video Generation Multimodal Video Generation
Nanjing University · University of Science and Technology of China · The Hong Kong University of Science and Technology (Guangzhou) +1
VA-Judger: Reward Modeling from Human Preference Feedback for Joint Video-Audio Generation
Aug 19, 2026Yinming Huang, Shuyuan Tu, Xi Yan +6Reward Modeling Audio-Video Generation
Fudan University · Shanghai Innovation Institution · Yinwang Intelligent Technology Co., Ltd +1
WorldJen: An End-to-End Multi-Dimensional Benchmark for Generative Video Models
May 5, 2026Karthik Inbasekar, Guy Rom, Omer ShlomovitsHuman Preference Evaluation Video Quality Assessment

Oct 15, 2026J/K move · Enter open · S save

ProactGen: Learning Proactive Clarification for Video Generation

Zhixiao Wang, Yang Chen, Yifan Xu, Chenbin Li+4

Nanjing University · University of Science and Technology of China · The Hong Kong University of Science and Technology (Guangzhou)+1

Video Generation Multimodal Video Generation Agentic Video Generation

Open paper AskPDF arXiv

Recent advances in video generation let users create high-quality videos from natural language descriptions, yet obtaining results that match user intent remains time-consuming: individual generations are computationally expensive, and underspecified prompts force repeated cycles of generation, inspection, and revision. Existing video prompt optimization methods improve generation through automatic rewriting, but cannot supply the ideas and preferences that were not in the initial description. We propose ProactGen, a video generation framework that learns proactive clarification, eliciting user intent through questions grounded in generated video drafts. Specifically, we train a vision-language model as a questioning policy with reinforcement learning: observing the initial prompt and a video draft, it learns to ask the questions most valuable to downstream generation, rewarded by the alignment between the final generated video and a target reference. We further find that effective questioning does not require drafts of final-output quality, and exploit this property by generating drafts with fewer denoising steps, keeping the added cost of clarification low. Experiments show that ProactGen improves target alignment over direct generation by at least 1.8x the gain of the strongest baselines. Moreover, a single round of ProactGen already exceeds the alignment that the strongest baseline reaches after four refinement rounds, in 29% of its time.

Abstract

Text-to-Audio-Video Generation Evaluation

Explore similar work

ProactGen: Learning Proactive Clarification for Video Generation

VA-Judger: Reward Modeling from Human Preference Feedback for Joint Video-Audio Generation

WorldJen: An End-to-End Multi-Dimensional Benchmark for Generative Video Models