{"ID":2850872,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.22045","arxiv_id":"2510.22045","title":"VLM-SlideEval: Evaluating VLMs on Structured Comprehension and Perturbation Sensitivity in PPT","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.","short_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 VLM...","url_abs":"https://arxiv.org/abs/2510.22045","url_pdf":"https://arxiv.org/pdf/2510.22045v1","authors":"[\"Hyeonsu Kang\",\"Emily Bao\",\"Anjan Goswami\"]","published":"2025-10-24T22:06:56Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Language Model\"]","has_code":false}