Elamur Research⌕
☰
home / papers / 2025 / PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slides

PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slides

Paper · 2025 · Hao Zheng, Xinyan Guan, Hao Kong et al.

Summary

Most automatic slide generators treat the task as text-to-slides: they summarize a document and pour the text into fixed templates or rules. The resulting decks tend to be wordy, visually dull, and poorly structured. PPTAgent instead imitates how people usually work, by adapting existing well-designed slides. It runs in two stages:

  • Stage I (analysis): an LLM/MLLM pipeline studies a reference presentation. It separates structural slides from content slides, clusters the content slides by layout, and extracts a content schema for each cluster.
  • Stage II (generation): the model drafts an outline that links each new slide to a purpose, a source section, and a reference slide. It then emits executable editing actions (e.g., replace_span) through APIs that operate on HTML-rendered slides. When an action fails, the error is fed back so the model can correct itself.

The authors also propose PPTEval, an MLLM-as-a-judge framework that scores decks on Content, Design, and Coherence. PPTAgent outperforms prior generators on all three dimensions. This matters because it shifts presentation generation toward reusing references and evaluating decks holistically, not just checking text quality.

Key points

  • PPTAgent averages 3.67 across PPTEval's three dimensions. Its generation success rate on the authors' dataset exceeds 95%.
  • PPTEval ratings correlate with human judgments at an average Pearson of 0.71.
  • Traditional metrics correlate only weakly with PPTEval. For example, KCTV leads on ROUGE-L (16.76) and PPL (41.41) yet scores only 2.55 on PPTEval Content. PPTAgent scores 3.28 on Content despite a ROUGE-L of 14.25 and PPL of 496.62.
  • The authors release a new dataset, Zenodo10K, built from license-filtered Zenodo presentations, along with code at github.com/icip-cas/PPTAgent.
  • Stated limitations include occasional generation failures, a weak retrieval component, and limited use of visual cues for fine layout fixes such as preventing overlapping elements.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🏛️Chinese Academy of SciencesBeijing · China194 works · 23,906 citations🎓University of Chinese Academy of SciencesBeijing · China147 works · 8,480 citations🔬Institute of SoftwareBeijing · China21 works · 4,980 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksComputer Vision and Pattern Recognition · 2,869 works

Cite

Cite as Hao Zheng, Xinyan Guan, Hao Kong, Wenkai Zhang, Jia Zheng, Weixiang Zhou, Hongyu Lin, Yaojie Lu, Xianpei Han and Le Sun. 2025. PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slides. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2025.emnlp-main.728

@inproceedings{zheng2025pptagent,
  title = {{PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slides}},
  author = {Hao Zheng, Xinyan Guan, Hao Kong, Wenkai Zhang, Jia Zheng, Weixiang Zhou, Hongyu Lin, Yaojie Lu, Xianpei Han and Le Sun},
  booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing},
  year = {2025},
  doi = {10.18653/v1/2025.emnlp-main.728},
  url = {https://doi.org/10.18653/v1/2025.emnlp-main.728}
}

References

38 references, 13 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Sambaran Bandyopadhyay, Himanshu Maheshwari, Anandhavelu Natarajan, and Apoorv Saxena. 2024. find
  2. Enhancing presentation slide generation by llms with a multi-staged end-to-end approach. arXiv preprint arXiv:2406.06556. find
  3. Andrea Barrick, Dana Davis, and Dana Winkler. 2018. Image versus text in powerpoint lectures: Who does it benefit? Journal of Baccalaureate Social Work, 23(1):91–109. find
  4. Isabel Alyssa Cachola, Silviu Cucerzan, Allen Herring, Vuksan Mijovic, Erik Oveson, and Sujay Kumar Jauhar. 2024. Knowledge-centric templatic views of documents. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 15460–15476, Miami, Florida, USA. Association for Computat… find
  5. Dongping Chen, Ruoxi Chen, Shilin Zhang, Yinuo Liu, Yaochen Wang, Huichi Zhou, Qihui Zhang, Pan Zhou, Yao Wan, and Lichao Sun. 2024a. Mllm-as-a-judge: Assessing multimodal llm-as-a-judge with vision-language benchmark. arXiv preprint arXiv:2402.04788.
  6. Jianlv Chen, Shitao Xiao, Peitian Zhang, Kun Luo, Defu Lian, and Zheng Liu. 2024b. Bge m3-embedding: Multi-lingual, multi-functionality, multi-granularity text embeddings through self-knowledge distillation. arXiv preprint arXiv:2402.03216.
  7. Jiawei Chen, Xinyan Guan, Qianhao Yuan, Guozhao Mo, Weixiang Zhou, Yaojie Lu, Hongyu Lin, Ben He, Le Sun, and Xianpei Han. 2025. Consistentchat: Building skeleton-guided consistent dialogues for large language models from scratch. arXiv preprint arXiv:2506.03558. — [arXiv:2506.03558](https://arxiv.org/abs/2506.03558) find
  8. Xiang Deng, Yu Gu, Boyuan Zheng, Shijie Chen, Sam Stevens, Boshi Wang, Huan Sun, and Yu Su. 2024. Mind2web: Towards a generalist agent for the web. Advances in Neural Information Processing Systems, 36.
  9. Nancy Duarte. 2008. Slide: ology: The art and science of creating great presentations, volume 1. O’Reilly Media Sebastapol. find
  10. Nancy Duarte. 2010. Resonate: Present visual stories that transform audiences. John Wiley & Sons. find
  11. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783.
  12. European Organization For Nuclear Research and OpenAIRE. 2013. Zenodo. find
  13. Weixi Feng, Wanrong Zhu, Tsu-jui Fu, Varun Jampani, Arjun Akula, Xuehai He, Sugato Basu, Xin Eric Wang, and William Yang Wang. 2024. Layoutgpt: Compositional visual planning and generation with large language models. Advances in Neural Information Processing Systems, 36. find
  14. Tsu-Jui Fu, William Yang Wang, Daniel McDuff, and Yale Song. 2022. Doc2ppt: Automatic presentation slides generation from scientific documents. Proceedings of the AAAI Conference on Artificial Intelligence, 36(1):634–642. find
  15. Jiaxin Ge, Zora Zhiruo Wang, Xuhui Zhou, Yi-Hao Peng, Sanjay Subramanian, Qinyue Tan, Maarten Sap, Alane Suhr, Daniel Fried, Graham Neubig, et al. 2025. Autopresent: Designing structured visuals from scratch. arXiv preprint arXiv:2501.00912. find
  16. Michael Robert Gryk. 2022. Human readability of data files. Balisage series on markup technologies, 27. find
  17. Xinyan Guan, Yanjiang Liu, Hongyu Lin, Yaojie Lu, Ben He, Xianpei Han, and Le Sun. 2024. Mitigating large language model hallucinations via autonomous knowledge graph-based retrofitting. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 18126–18134. find
  18. Yiduo Guo, Zekai Zhang, Yaobo Liang, Dongyan Zhao, and Duan Nan. 2023. Pptc benchmark: Evaluating large language models for powerpoint task completion. arXiv preprint arXiv:2311.01767. find
  19. Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. 2017. Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30. find
  20. Ryo Kamoi, Yusen Zhang, Nan Zhang, Jiawei Han, and Rui Zhang. 2024. When can llms actually correct their own mistakes? a critical survey of self-correction of llms. Transactions of the Association for Computational Linguistics, 12:1417–1440.
  21. Wai-Chung Kwan, Xingshan Zeng, Yuxin Jiang, Yufei Wang, Liangyou Li, Lifeng Shang, Xin Jiang, Qun Liu, and Kam-Fai Wong. 2024. Mt-eval: A multi-turn capabilities evaluation benchmark for large language models. Preprint, arXiv:2401.16745.
  22. Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph Gonzalez, Hao Zhang, and Ion Stoica. 2023. Efficient memory management for large language model serving with pagedattention. In Proceedings of the 29th Symposium on Operating Systems Principles, pages 611–626.
  23. Md Tahmid Rahman Laskar, Sawsan Alqahtani, M Saiful Bari, Mizanur Rahman, Mohammad Abdullah Matin Khan, Haidar Khan, Israt Jahan, Amran Bhuiyan, Chee Wei Tan, Md Rizwan Parvez, Enamul Hoque, Shafiq Joty, and Jimmy Huang. 2024. A systematic survey and critical review on evaluating large language mode…
  24. Yanda Li, Chi Zhang, Wanqi Yang, Bin Fu, Pei Cheng, Xin Chen, Ling Chen, and Yunchao Wei. 2024. Appagent v2: Advanced agent for flexible mobile interactions. arXiv preprint arXiv:2408.11824. — [arXiv:2408.11824](https://arxiv.org/abs/2408.11824) find
  25. Chin-Yew Lin. 2004. ROUGE: A package for automatic evaluation of summaries. In Text Summarization Branches Out, pages 74–81, Barcelona, Spain. Association for Computational Linguistics.
  26. Yang Liu, Dan Iter, Yichong Xu, Shuohang Wang, Ruochen Xu, and Chenguang Zhu. 2023. G-eval: NLG evaluation using gpt-4 with better human alignment. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 2511–2522, Singapore. Association for Computational Lin…
  27. Himanshu Maheshwari, Sambaran Bandyopadhyay, Aparna Garimella, and Anandhavelu Natarajan. 2024. Presentations are not always linear! gnn meets llm for document-to-presentation transformation with attribution. arXiv preprint arXiv:2405.13095. 2020. Visual transformers: Token-based image representatio… — [arXiv:2405.13095](https://arxiv.org/abs/2405.13095) find
  28. Tong Wu, Guandao Yang, Zhibing Li, Kai Zhang, Ziwei Liu, Leonidas Guibas, Dahua Lin, and Gordon Wetzstein. 2024. Gpt-4v (ision) is a human-aligned evaluator for text-to-3d generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22227–22238. find
  29. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, et al. 2024. Qwen2. 5 technical report. arXiv preprint arXiv:2412.15115. find
  30. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric Xing, et al. 2023. Judging llm-as-a-judge with mt-bench and chatbot arena. Advances in Neural Information Processing Systems, 36:46595–46623.
  31. Ishani Mondal, S Shwetha, Anandhavelu Natarajan, Aparna Garimella, Sambaran Bandyopadhyay, and Jordan Boyd-Graber. 2024. Presentations by the humans and for the humans: Harnessing llms for generating persona-aware slides from documents. In Proceedings of the 18th Conference of the European Chapter o… find
  32. Athar Sefid, Prasenjit Mitra, and Lee Giles. 2021. Slidegen: an abstractive section-based slide generator for scholarly documents. In Proceedings of the 21st ACM Symposium on Document Engineering, pages 1–4. find
  33. Edward Sun, Yufang Hou, Dakuo Wang, Yunfeng Zhang, and Nancy XR Wang. 2021. D2s: Document-to-slide generation via query-based text summarization. arXiv preprint arXiv:2105.03664. find
  34. Hao Tang, Darren Key, and Kevin Ellis. 2025. Worldcoder, a model-based llm agent: Building world models by writing code and interacting with the environment. Advances in Neural Information Processing Systems, 37:70148–70212. find
  35. VikParuchuri. 2023. marker. find
  36. Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, et al. 2024a. Qwen2-vl: Enhancing vision-language model’s perception of the world at any resolution. arXiv preprint arXiv:2409.12191.
  37. Xingyao Wang, Yangyi Chen, Lifan Yuan, Yizhe Zhang, Yunzhu Li, Hao Peng, and Heng Ji. 2024b. Executable code actions elicit better llm agents. arXiv preprint arXiv:2402.01030.
  38. Bichen Wu, Chenfeng Xu, Xiaoliang Dai, Alvin Wan, Peizhao Zhang, Zhicheng Yan, Masayoshi Tomizuka, Joseph Gonzalez, Kurt Keutzer, and Peter Vajda. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/zheng-2025-pptagent-generating-evaluating-presentations-beyond/