Improve dataset card for Zenodo10K: Add VLM-SlideEval paper context, task category, and license

README.md

@@ -1,4 +1,7 @@
 ---
+task_categories:
+- image-text-to-text
+license: other
 dataset_info:
   features:
   - name: filename
@@ -34,10 +37,15 @@ configs:
 
 # PPTAgent/Zenodo10K
 
-This is the dataset used in [PPTAgent](https://arxiv.org/abs/2501.03936), crawled from [zenodo](http://zenodo.org).
+This is the dataset used in [PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slides](https://arxiv.org/abs/2501.03936) and also used for evaluation in [VLM-SlideEval: Evaluating VLMs on Structured Comprehension and Perturbation Sensitivity in PPT](https://huggingface.co/papers/2510.22045). It was crawled from [zenodo](http://zenodo.org).
 To the best of our knowledge, it is the **largest presentation dataset** currently available, comprising over 10,000 **PowerPoint (.pptx)** files, all distributed under a clear and compliant license.
 
-For more information, please visit our [github repo](https://github.com/icip-cas/PPTAgent).
+For more information regarding the `PPTAgent` project, please visit its [github repo](https://github.com/icip-cas/PPTAgent).
+
+## VLM-SlideEval 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 ( this https URL ), 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.
+
+## Sample Usage
 
 ```python
 dirname = f"zenodo-pptx/pptx/{task['license']}/{task['created'][:4]}/"
@@ -62,5 +70,12 @@ If you find this project helpful, please use the following to cite it:
       url={https://arxiv.org/abs/2501.03936}, 
 }
 ```
-
-
+If you use this dataset for VLM evaluation, please also cite:
+```bibtex
+@article{xu2025vlm,
+  title={VLM-SlideEval: Evaluating VLMs on Structured Comprehension and Perturbation Sensitivity in PPT},
+  author={Xu, Charles and Li, Qiyang and Luo, Jianlan and Levine, Sergey},
+  journal={arXiv preprint arXiv:2510.22045},
+  year={2025}
+}
+```