A Good Talk Does not Look Like a Summary, It Teaches You! Measuring Takeaways from Paper-to-Video Talks
Not on the map. Why?
Abstract (arXiv)
Automatically generated videos from scientific papers are increasingly used for education and research dissemination. However, existing evaluation metrics mainly measure visual quality or whether key points from the paper appear in the video without assessing whether the video actually helps viewers understand the ideas. We introduce EffectivePresentationScorer, a framework for evaluating the instructional quality of scientific presentation videos. It checks whether a video explains the main ideas clearly, introduces needed background concepts, and connects technical details to the main contribution of the paper. When we apply EffectivePresentationScorer to the existing paper-to-video generation systems, we find that generated videos mention the correct topics and follow the structure of the paper but fail to explain prerequisite concepts or clarify why the method works. These failures are often ignored by existing video evaluation metrics, which focus on content presence rather than explanatory quality.
Key facts. No citing papers in the map yet (data through 13 Oct 2026). Builds on 21 of its 36 references that are in the map. First posted on arXiv 26 Jun 2026.
Builds on
21 of 36 references.
References · 21
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,588
- 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,107
- 2023VBench: Comprehensive Benchmark Suite for Video Generative Modelscited by 1,775
- 2018Towards Accurate Generative Models of Video: A New Metric & Challengescited by 1,217
- 2020Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillationcited by 591
- 2024Sora: A Review on Background, Technology, Limitations, and Opportunities of Large Vision Modelscited by 473
- 2023EvalCrafter: Benchmarking and Evaluating Large Video Generation Modelscited by 334
- 2024VideoPhy: Evaluating Physical Commonsense for Video Generationcited by 219
- 2025Video models are zero-shot learners and reasonerscited by 213
- 2023VIEScore: Towards Explainable Metrics for Conditional Image Synthesis Evaluationcited by 197
- 2024VideoScore: Building Automatic Metrics to Simulate Fine-grained Human Feedback for Video Generationcited by 190
- 2025PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slidescited by 84
9 more
- 2024Evaluation of Text-to-Video Generation Models: A Dynamics Perspectivecited by 66
- 2021D2S: Document-to-Slide Generation Via Query-Based Text Summarizationcited by 50
- 2024VBench++: Comprehensive and Versatile Benchmark Suite for Video Generative Modelscited by 43
- 2025VideoGen-Eval: Agent-based System for Video Generation Evaluationcited by 23
- 2021DefSent: Sentence Embeddings using Definition Sentencescited by 20
- 2025We Should Evaluate Real-World Impactcited by 12
- 2025What Is That Talk About? A Video-to-Text Summarization Dataset for Scientific Presentationscited by 11
- 2021Training Adaptive Computation for Open-Domain Question Answering with Computational Constraintscited by 4
- 2025LecEval: An Automated Metric for Multimodal Knowledge Acquisition in Multimedia Learningcited by 1
Built on by
Not cited yet.