VLM-SlideEval: Evaluating VLMs on Structured Comprehension and Perturbation Sensitivity in PPT

TL;DR

VLM-SlideEval is a comprehensive framework for assessing vision-language models' ability to understand, extract, and interpret structured content and narrative in presentation slides, revealing current limitations and guiding future improvements.

Contribution

The paper introduces VLM-SlideEval, a novel evaluation framework specifically designed for slide content understanding and robustness testing of VLMs.

Findings

VLMs struggle with pixel-accurate element extraction from slides.

VLMs maintain some robustness under controlled perturbations.

VLMs are less effective at understanding narrative structure across multiple slides.

Abstract

Vision-language models (VLMs) are increasingly used to evaluate multimodal content, including presentation slides, yet their slide-specific understanding remains underexplored {despite their growing role as critics in agentic, model-forward pipelines}. We introduce VLM-SlideEval, an evaluation framework that probes VLMs along three axes: (1) element-level extraction from slide images aligned to ground truth; (2) robustness to controlled perturbations in geometry, style, and text; and (3) higher-level comprehension, such as recovering a deck's narrative order from shuffled slides. Using publicly available decks from Zenodo (https://huggingface.co/datasets/Forceless/Zenodo10K/viewer/default/pptx), we standardize ground-truth element metadata from PowerPoint XML and live renderings into a unified, verifiable schema. Empirically, VLMs underperform on pixel-accurate extraction and show non-trivial agreement, fidelity, and consistency under controlled perturbations, while performing better on single-slide content understanding; however, they do not reliably capture narrative structure across slides. These results highlight the limits of current VLMs for slide evaluation and motivate calibrated, critic-in-the-loop evaluators that drive iterative refinement and selection in agentic pipelines.