Animation Needs Attention: A Holistic Approach to Slides Animation Comprehension with Visual-Language Models

TL;DR

This paper introduces a new dataset and model fine-tuning approach for improving AI understanding and generation of slide animations, addressing a key gap in visual-language models for dynamic presentation content.

Contribution

The authors release the first public dataset for slide-animation modeling and demonstrate that fine-tuning with Low-Rank Adaptation significantly enhances animation comprehension and generation capabilities of vision-language models.

Findings

LoRA fine-tuning improves BLEU-4 by 60% and ROUGE-L by 30%.

The CODA metric effectively evaluates animation action coverage, order, and detail.

The dataset enables better temporal reasoning in slide animation understanding.

Abstract

Slide animations, such as fade-in, fly-in, and wipe, are critical for audience engagement, efficient information delivery, and vivid visual expression. However, most AI-driven slide-generation tools still lack native animation support, and existing vision-language models (VLMs) struggle with animation tasks due to the absence of public datasets and limited temporal-reasoning capabilities. To address this gap, we release the first public dataset for slide-animation modeling: 12,000 triplets of natural-language descriptions, animation JSON files, and rendered videos, collectively covering every built-in PowerPoint effect. Using this resource, we fine-tune Qwen-2.5-VL-7B with Low-Rank Adaptation (LoRA) and achieve consistent improvements over GPT-4.1 and Gemini-2.5-Pro in BLEU-4, ROUGE-L, SPICE, and our Coverage-Order-Detail Assessment (CODA) metric, which evaluates action coverage, temporal order, and detail fidelity. On a manually created test set of slides, the LoRA model increases BLEU-4 by around 60%, ROUGE-L by 30%, and shows significant improvements in CODA-detail. This demonstrates that low-rank adaptation enables reliable temporal reasoning and generalization beyond synthetic data. Overall, our dataset, LoRA-enhanced model, and CODA metric provide a rigorous benchmark and foundation for future research on VLM-based dynamic slide generation.