Elamur Research⌕
☰
home / papers / 2026 / KCVR: Knowledge-Centric Video Reconstruction for Structured Pedagogical Summarization via Dynamic Graph Planning

KCVR: Knowledge-Centric Video Reconstruction for Structured Pedagogical Summarization via Dynamic Graph Planning

Paper · 2026 · Jingjiang Liu, Jia Zhu, Hanghui Guo et al.

Summary

Standard video summarizers aim at quick overviews. On lecture videos they often scramble the teaching logic, for example stating a proof before the definition it relies on. The authors define a new task, Structure-Pedagogical Reconstruction (SPR): producing a structured summary whose concept order follows prerequisites and is backed by what appears on the blackboard.

Their system, KCVR, plans first and writes second. A transcript-based parser (P³) divides the lecture into teaching phases. The SG-Planner then trims a Dual-Layer Epistemic Graph, removing concepts the video does not support. Generation uses two mechanisms:

  • KGVA links each active concept to blackboard regions.
  • ACP blocks concepts that should come later in the plan.

The authors also release EduStruct, a benchmark spanning 10 disciplines, together with structure-oriented metrics (KPC, LOC). KCVR beats end-to-end Video-LLM baselines and stays robust when the transcript is noisy or missing. This suggests that explicit structural priors help make educational summaries trustworthy.

Key points

  • EduStruct benchmark: more than 460 videos across 10 disciplines, paired with graphs that separate the curriculum layer (TeachingPoint) from the concept layer (KnowledgeConcept). Edges include NEXT, BASED_ON, PREREQ_OF and HAS_CONCEPT.
  • Noisy transcripts: under ASR noise, KCVR scores 61.9 KPC / 59.3 LOC, compared with 54.9 / 46.0 for InternVL3 and 57.8 / 50.5 for Text-RAG. This is only a 7.5% drop from in-domain performance.
  • Visual-only input (8 frames): KCVR keeps 55.5 KPC, against 42.0 for InternVL3. The robustness gains also appear on a VideoLLaMA2 backbone.
  • Planning is the bottleneck: substituting an oracle plan adds +3.41 KPC and +2.99 LOC. The planner itself adds less than 5% latency.
  • Domains without a graph: on InfoTech without a KG, KPC falls from 66.79 to 59.67, about a 10.7% drop.

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

🎓Hong Kong University of Science and TechnologyHong Kong · China142 works · 25,136 citations🏢Alibaba GroupHangzhou · China168 works · 15,386 citations🎓Soochow UniversitySuzhou · China63 works · 4,507 citations🎓Southeast UniversityNanjing · China28 works · 354 citations🎓Southeast UniversityDhaka · Bangladesh8 works · 160 citations🎓Zhejiang Normal UniversityJinhua · China4 works · 16 citations

 Fields, topics and concepts

📚 Shelves 2The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksComputer Vision and Pattern Recognition · 2,869 works

🧭 Topics 1Research subjects and languages the work is about.Summarization · 232 works

Cite

Cite as Jingjiang Liu, Jia Zhu, Hanghui Guo, Weijie Shi, Yue Cui, Xiaokang Jin, Yilin Wang, Qingyu Niu, Jiawei Shen, Guoqing Ma, Yidan Liang, Shimin Di and Jiajie Xu. 2026. KCVR: Knowledge-Centric Video Reconstruction for Structured Pedagogical Summarization via Dynamic Graph Planning. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.414

@inproceedings{liu2026kcvr,
  title = {{KCVR: Knowledge-Centric Video Reconstruction for Structured Pedagogical Summarization via Dynamic Graph Planning}},
  author = {Jingjiang Liu, Jia Zhu, Hanghui Guo, Weijie Shi, Yue Cui, Xiaokang Jin, Yilin Wang, Qingyu Niu, Jiawei Shen, Guoqing Ma, Yidan Liang, Shimin Di and Jiajie Xu},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.414},
  url = {https://doi.org/10.18653/v1/2026.acl-long.414}
}

References

38 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Kevin Ackermans, Björn B. de Koning, and Halszka Jarodzka. 2025. Instructional videos and deeper processing: Insights and applications. Learning and Instruction, 98:102137. find
  2. Evlampios Apostolidis, Eleni Adamantidou, Alexandros I. Metsai, Vasileios Mezaris, and Ioannis Patras. 2021. AC-SUM-GAN: Connecting actor-critic and generative adversarial networks for unsupervised video summarization. IEEE Transactions on Circuits and Systems for Video Technology, 31(8):3278–3292. find
  3. Fatemeh Bagheri, Anshuman Garga, and Ramon E. Lopez. 2024. Exploring radio emissions from confirmed exoplanets using SKA. Preprint, arXiv:2404.14468. — [arXiv:2404.14468](https://arxiv.org/abs/2404.14468) find
  4. Z. Chen and 1 others. 2024. Intern VL: Scaling up vision foundation models and aligning for generic visual-linguistic tasks. In Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 24185–24198. find
  5. Chaoyou Fu, Yuhan Dai, Yongdong Luo, Lei Li, Shuhuai Ren, Renrui Zhang, Zihan Wang, Chenyu Zhou, Yunhang Shen, Mengdan Zhang, Peixian Chen, Yanwei Li, Shaohui Lin, Sirui Zhao, Ke Li, Tong Xu, Xiawu Zheng, Enhong Chen, Caifeng Shan, and 2 others. 2025. Video-mme: The first-ever comprehensive evaluati…
  6. Lishuai Gao, Yujie Zhong, Yingsen Zeng, Haoxian Tan, Dengjie Li, and Zheng Zhao. 2024. Linvt: Empower your image-level large language model to understand videos. Preprint, arXiv:2412.05185. — [arXiv:2412.05185](https://arxiv.org/abs/2412.05185) find
  7. Hanghui Guo, Jia Zhu, Shimin Di, Weijie Shi, Zhangze Chen, and Jiajie Xu. 2025. DioR: Adaptive cognitive detection and contextual retrieval optimization for dynamic retrieval-augmented generation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: L… find
  8. Haoyu Han, Yu Wang, Harry Shomer, Kai Guo, Jiayuan Ding, Yongjia Lei, Mahantesh Halappanavar, Ryan A. Rossi, Subhabrata Mukherjee, Xianfeng Tang, Qi He, Zhigang Hua, Bo Long, Tong Zhao, Neil Shah, Amin Javari, Yinglong Xia, and Jiliang Tang. 2025. Retrieval-augmented generation with graphs (GraphRAG… find
  9. Bo He, Hengduo Li, Young Kyun Jang, Menglin Jia, Xuefei Cao, Ashish Shah, Abhinav Shrivastava, and Ser-Nam Lim. 2024. Ma-lmm: Memory-augmented large multimodal model for long-term video understanding. Preprint, arXiv:2404.05726. find
  10. Weifeng Hu, Yu Zhang, Yujun Li, Jia Zhao, Xifeng Hu, Yan Cui, and Xuejing Wang. 2023. Query-based video summarization with multi-label classification network. Multimedia Tools Appl., 82(24):37529– 37549. find
  11. Hang Hua, Yolo Yunlong Tang, Chenliang Xu, and Jiebo Luo. 2025. V2Xum-LLM: Cross-modal video summarization with temporal prompt instruction tuning. Preprint, arXiv:2404.12353. — [arXiv:2404.12353](https://arxiv.org/abs/2404.12353) find
  12. Ximing Lu, Sean Welleck, Peter West, Liwei Jiang, Jungo Kasai, Daniel Khashabi, Ronan Le Bras, Lianhui Qin, Youngjae Yu, Rowan Zellers, Noah A. Smith, and Yejin Choi. 2022. NeuroLogic a*esque decoding: Constrained text generation with lookahead heuristics. In Proceedings of the 2022 Conference of th…
  13. Zhisheng Huang, Xudong Jia, Tao Chen, and Zhongwei Zhang. 2025. A general scalable approach for knowledge selection based on iterative hybrid encoding and re-ranking. Data Intelligence, 7(1):124–142. find
  14. Ximing Lu, Peter West, Rowan Zellers, Ronan Le Bras, Chandra Bhagavatula, and Yejin Choi. 2021. Neuro-Logic decoding: (un)supervised neural text generation with predicate logic constraints. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Lingu… find
  15. Peng Jin, Ryuichi Takanobu, Wancai Zhang, Xiaochun Cao, and Li Yuan. 2024. Chat-UniVi: Unified visual representation empowers large language models with image and video understanding. Preprint, arXiv:2311.08046. — [arXiv:2311.08046](https://arxiv.org/abs/2311.08046) find
  16. Linhao Luo, Zicheng Zhao, Gholamreza Haffari, Yuan-Fang Li, Chen Gong, and Shirui Pan. 2025. Graph-constrained Reasoning: Faithful reasoning on knowledge graphs with large language models. Preprint, arXiv:2410.13080. — [arXiv:2410.13080](https://arxiv.org/abs/2410.13080) find
  17. Prem Nigam Kar, David E. Roberson, Tim Seppelt, and Peter Zeman. 2025. NPA hierarchy for quantum isomorphism and homomorphism indistinguishability. Leibniz International Proceedings in Informatics, 334:105:1–105:19. find
  18. Muhammad Maaz, Hanoona Rasheed, Salman Khan, and Fahad Shahbaz Khan. 2024. Video-ChatGPT: Towards detailed video understanding via large vision and language models. Preprint, arXiv:2306.05424.
  19. Pei Ke, Haozhe Ji, Yu Ran, Xin Cui, Liwei Wang, Linfeng Song, Xiaoyan Zhu, and Minlie Huang. 2021. Jointgt: Graph-text joint representation learning for text generation from knowledge graphs. Preprint, arXiv:2106.10502. find
  20. Shehan Munasinghe, Rusiru Thushara, Muhammad Maaz, Hanoona Abdul Rasheed, Salman Khan, Mubarak Shah, and Fahad Khan. 2023. Pg-video-llava: Pixel grounding large video-language models. Preprint, arXiv:2311.13435. — [arXiv:2311.13435](https://arxiv.org/abs/2311.13435) find
  21. Shuailin Li, Yuang Zhang, Yucheng Zhao, Qiuyue Wang, Fan Jia, Yingfei Liu, and Tiancai Wang. 2023. Vlm-eval: A general evaluation on video large language models. Preprint, arXiv:2311.11865. — [arXiv:2311.11865](https://arxiv.org/abs/2311.11865) find
  22. Medhini Narasimhan, Anna Rohrbach, and Trevor Darrell. 2021. CLIP-It!: Language-guided video summarization. In Proceedings of the 35th International Conference on Neural Information Processing Systems. find
  23. Jingyang Lin, Hang Hua, Ming Chen, Yikang Li, Jenhao Hsiao, Chiuman Ho, and Jiebo Luo. 2024. VideoXum: Cross-modal visual and textural summarization of videos. IEEE Transactions on Multimedia, 26:5548–5560. find
  24. Torbjørn Netland, Oliver von Dzengelevski, Katalin Tesch, and Daniel Kwasnitschka. 2025. Comparing human-made and AI-generated teaching videos: An experimental study on learning effects. Comput. Educ., 224(C). find
  25. Dongqi Liu, Chenxi Whitehouse, Xi Yu, Louis Mahon, Rohit Saxena, Zheng Zhao, Yifu Qiu, Mirella Lapata, and Vera Demberg. 2025. What is that talk about? a video-to-text summarization dataset for scientific presentations. In Proceedings of the 63rd Annual Meeting of the Association for Computational L…
  26. Shuhuai Ren, Linli Yao, Shicheng Li, Xu Sun, and Lu Hou. 2024. TimeChat: A time-sensitive multimodal large language model for long video understanding. In Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 14313–14323. find
  27. Yuanxin Liu, Shicheng Li, Yi Liu, Yuxiang Wang, Shuhuai Ren, Lei Li, Sishuo Chen, Xu Sun, and Lu Hou. 2024. TempCompass: Do video LLMs really understand videos? In Findings of the Association for Computational Linguistics: ACL 2024, pages 8731–8772, Bangkok, Thailand. Association for Computational L… find
  28. Enxin Song, Wenhao Chai, Guanhong Wang, Yucheng Zhang, Haoyang Zhou, Feiyang Wu, Haozhe Chi, Xun Guo, Tian Ye, Yanting Zhang, Yan Lu, Jenq-Neng Hwang, and Gaoang Wang. 2024. MovieChat: From dense token to sparse memory for long video understanding. Preprint, arXiv:2307.16449. find
  29. Yolo Y. Tang, Jing Bi, Siting Xu, Luchuan Song, Susan Liang, Teng Wang, Daoan Zhang, Jie An, Jingyang Lin, Rongyi Zhu, Ali Vosoughi, Chao Huang, Zeliang Zhang, Pinxin Liu, Mingqian Feng, Feng Zheng, Jianguo Zhang, Ping Luo, Jiebo Luo, and Chenliang Xu. 2025. Video understanding with large language m… find
  30. Lifu Tu, Semih Yavuz, Jin Qu, Jiacheng Xu, Rui Meng, Caiming Xiong, and Yingbo Zhou. 2024. Unlocking anticipatory text generation: A constrained approach for large language models decoding. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 15532–15548,… find
  31. Shuo Wang and Jihao Zhang. 2025. MF2Summ: Multimodal fusion for video summarization with temporal alignment. Preprint, arXiv:2506.10430. — [arXiv:2506.10430](https://arxiv.org/abs/2506.10430) find
  32. Peng Xiao, Chao Liu, Wei Jia, and Lijun Dong. 2025. Aligned-entities-based fusion embedding on hetero-field knowledge graphs. Data Intelligence, 7(3):618– 635. find
  33. Tao Xu, Yuan Liu, Yaru Jin, Yueyao Qu, Jie Bai, Wenlan Zhang, and Yun Zhou. 2025. From recorded to AI-generated instructional videos: A comparison of learning performance and experience. British Journal of Educational Technology, 56(4):1463–1487. find
  34. Qinghao Ye, Haiyang Xu, Guohai Xu, Jiabo Ye, Ming Yan, Yiyang Zhou, Junyang Wang, Anwen Hu, Pengcheng Shi, Yaya Shi, Chenliang Li, Yuanhong Xu, Hehong Chen, Junfeng Tian, Qi Qian, Ji Zhang, Fei Huang, and Jingren Zhou. 2024. mPLUG-Owl: Modularization empowers large language models with multimodality…
  35. Hang Zhang, Xin Li, and Lidong Bing. 2023. Video-LLaMA: An instruction-tuned audio-visual language model for video understanding. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 543–553, Singapore. Association for Computational…
  36. Lu Zhang, Tiancheng Zhao, Heting Ying, Yibo Ma, and Kyusong Lee. 2024. Omagent: A multimodal agent framework for complex video understanding with task divide-and-conquer. Preprint, arXiv:2406.16620. — [arXiv:2406.16620](https://arxiv.org/abs/2406.16620) find
  37. Jia Zhu, Hanghui Guo, Weijie Shi, Zhangze Chen, and Pasquale De Meo. 2025. RaDIO: Real-time hallucination detection with contextual index optimized query formulation for dynamic retrieval-augmented generation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 26129–… find
  38. Wencheng Zhu, Jiwen Lu, Jiahao Li, and Jie Zhou. 2021. Dsnet: A flexible detect-to-summarize network for video summarization. IEEE Transactions on Image Processing, 30:948–962. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/liu-2026-kcvr-knowledge-centric-video-reconstruction/