Elamur Research⌕
☰
home / papers / 2026 / CSI: An Investigative Multi-Agent Framework for Explainable Short Video Fake News Detection

CSI: An Investigative Multi-Agent Framework for Explainable Short Video Fake News Detection

Paper · 2026 · Yuxin Wang, Yang Yang and Huaiwen Zhang

Summary

The paper addresses detecting fake news in short videos (TikTok/Kuaishou-style content) while also explaining each decision. Black-box multimodal small models classify well but cannot justify their outputs. Prompted multimodal LLMs can produce explanations, but they tend to hallucinate when modalities conflict and reason from a single perspective. The authors propose CSI, a multi-agent system modeled on a human investigation team. In the Multimodal Forensics Unit, separate MLLM agents analyze video shots, audio, and web-retrieved evidence, then merge their findings into a master casefile. In the Case Review Unit, reviewer agents debate that casefile to reach a review result. A trainable Adjudicator uses several attention mechanisms to combine the casefile and review with the original news, which avoids fine-tuning the MLLMs. On the FakeSV and FakeTT datasets, CSI reaches state-of-the-art accuracy, runs faster and uses fewer tokens than strong prompted MLLM baselines, and produces explanations that human evaluators rated. This makes automated fact-checking of video content more trustworthy.

Key points

  • Preprocessing and visual analysis: Videos are split into shots with TransNetV2. The Visual Analyst covers five aspects: shot type, visual content, subtitles, footage source, and technical authenticity.
  • Audio and evidence agents: The Acoustic Analyst covers voice, transcript, BGM, and sound effects. The Intelligence Analyst turns the title and on-screen text into an event-caption query, then re-ranks search results by source authority (official institutions first, then international media, then domestic media).
  • Accuracy: CSI scores 90.77 on FakeSV and 89.63 on FakeTT. The best baseline, Gemini-2.5-Pro with CoT and Web-RAG, scores 83.03 and 81.93.
  • Efficiency: Across 5,615 videos, CSI averages 16.2 s and 8.7K tokens per sample, compared with 32.7 s and 11.1K tokens for Gemini-2.5-Pro.
  • Limitation and code: The framework struggles with breaking news that has no credible evidence online yet. Code is available at github.com/VFCenter/CSI.

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

🎓Inner Mongolia UniversityHohhot · China3 works · 1 citation

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksComputer Vision and Pattern Recognition · 2,869 worksSocial Sciences · 1,772 worksSociology and Political Science · 577 works

🧭 Topics 1Research subjects and languages the work is about.Vision & language · 657 works

Cite

Cite as Yuxin Wang, Yang Yang and Huaiwen Zhang. 2026. CSI: An Investigative Multi-Agent Framework for Explainable Short Video Fake News Detection. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.1274

@inproceedings{wang2026csi,
  title = {{CSI: An Investigative Multi-Agent Framework for Explainable Short Video Fake News Detection}},
  author = {Yuxin Wang, Yang Yang and Huaiwen Zhang},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1274},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1274}
}

References

50 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Marah Abdin, Jyoti Aneja, Harkirat Behl, Sébastien Bubeck, Ronen Eldan, Suriya Gunasekar, Michael Harrison, Russell J. Hewett, Mojan Javaheripi, Piero Kauffmann, James R. Lee, Yin Tat Lee, Yuanzhi Li, Weishung Liu, Caio C. T. Mendes, Anh Nguyen, Eric Price, Gustavo de Rosa, Olli Saarikivi, and 8 oth… — Phi-4 Technical Report
  2. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, Humen Zhong, Yuanzhi Zhu, Mingkun Yang, Zhaohai Li, Jianqiang Wan, Pengfei Wang, Wei Ding, Zheren Fu, Yiheng Xu, and 8 others. 2025. Qwen2.5-vl technical report. Preprint, arXiv:2502.13…
  3. Álvaro Barbosa and Kristine Dizon. 2020. The film sound analysis framework: A conceptual tool to interpret the cinematic experience. Journal of Science and Technology of the Arts, 12:81–96. find
  4. Fuqing Bie, Shiyu Huang, Xijia Tao, Zhiqin Fang, Leyi Pan, Junzhe Chen, Min Ren, Liuyu Xiang, and Zhaofeng He. 2025. Omniplay: Benchmarking omni-modal models on omni-modal game playing. Preprint, arXiv:2508.04361. — [arXiv:2508.04361](https://arxiv.org/abs/2508.04361) find
  5. Yuyan Bu, Qiang Sheng, Juan Cao, Peng Qi, Danding Wang, and Jintao Li. 2024. Fakingrecipe: Detecting fake news on short video platforms from the perspective of creative process. In Proceedings of the 32nd ACM International Conference on Multimedia, page 1351–1360. find
  6. Hyewon Choi and Youngjoong Ko. 2021. Using topic modeling and adversarial neural networks for fake news video detection. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management, page 2950–2954. find
  7. Zili Wang, Steven Ka Shing Yau, Zijuan Lin, Liyang Zhou, Chenyu Ran, Lingfeng Xiao, Chenglin Wu, and Jürgen Schmidhuber. 2024. Metagpt: Meta programming for A multi-agent collaborative framework. In ICLR 2024, Vienna, Austria, May 7-11, 2024.
  8. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, Luke Marris, Sam Petulla, Colin Gaffney, Asaf Aharoni, Nathan Lintz, Tiago Cardal Pais, Henrik Jacobsson, Idan Szpektor, Nan-Jiang Jiang, and 3416 other…
  9. Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, and Abdelrahman Mohamed. 2021. Hubert: Self-supervised speech representation learning by masked prediction of hidden units. IEEE/ACM Trans. Audio, Speech and Lang. Proc., 29:3451–3460.
  10. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2020. Unsupervised cross-lingual representation learning at scale. In Proceedings of the 58th Annual Meeting of the Association fo… find
  11. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics, pages 4171–4186.
  12. Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. 2021. An image is worth 16x16 words: Transformers for image recognition at scale. In ICLR 2…
  13. Mingfei Han, Linjie Yang, Xiaojun Chang, Lina Yao, and Heng Wang. 2025. Shot2story: A new benchmark for comprehensive understanding of multi-shot videos. In ICLR 2025, Singapore, April 24-28, 2025. find
  14. Shawn Hershey, Sourish Chaudhuri, Daniel P. W. Ellis, Jort F. Gemmeke, Aren Jansen, R. Channing Moore, Manoj Plakal, Devin Platt, Rif A. Saurous, Bryan Seybold, Malcolm Slaney, Ron J. Weiss, and Kevin Wilson. 2017. Cnn architectures for large-scale audio classification. In 2017 IEEE International Co… find
  15. Rongpei Hong, Jian Lang, Jin Xu, Zhangtao Cheng, Ting Zhong, and Fan Zhou. 2025. Following clues, approaching the truth: Explainable micro-video rumor detection via chain-of-thought reasoning. In Proceedings of the ACM on Web Conference 2025, page 4684–4698. find
  16. Sirui Hong, Mingchen Zhuge, Jonathan Chen, Xiawu Zheng, Yuheng Cheng, Jinlin Wang, Ceyao Zhang, Beizhe Hu, Qiang Sheng, Juan Cao, Yuhui Shi, Yang Li, Danding Wang, and Peng Qi. 2024. Bad actor, good advisor: Exploring the role of large language models in fake news detection. Proceedings of the AAAI… find
  17. Edward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. Lora: Low-rank adaptation of large language models. In ICLR 2022, Virtual Event, April 25-29, 2022.
  18. Diederik P. Kingma and Jimmy Ba. 2015. Adam: A method for stochastic optimization. In ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings.
  19. Mojtaba Komeili, Kurt Shuster, and Jason Weston. 2022. Internet-augmented dialogue generation. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics, pages 8460–8478.
  20. Klaus Krippendorff. 2011. Computing krippendorff’s alpha-reliability. Technical report, University of Pennsylvania, Annenberg School for Communication. Postprint version. find
  21. Manling Li, Ruochen Xu, Shuohang Wang, Luowei Zhou, Xudong Lin, Chenguang Zhu, Michael Zeng, Heng Ji, and Shih-Fu Chang. 2022. Clip-event: Connecting text and images with event structures. In CVPR 2022, New Orleans, LA, USA, June 18-24, 2022, pages 16399–16408. find
  22. 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.
  23. Yuhan Liu, Yuxuan Liu, Xiaoqing Zhang, Xiuying Chen, and Rui Yan. 2025. The Truth Becomes Clearer Through Debate! multi-agent systems with large language models unmask fake news. In Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR… find
  24. Shervin Minaee, Tomas Mikolov, Narjes Nikzad, Meysam Chenaghlu, Richard Socher, Xavier Amatriain, and Jianfeng Gao. 2025. Large language models: A survey. Preprint, arXiv:2402.06196.
  25. OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff Belgum, and 262 others.… — GPT-4 Technical Report
  26. Joon Sung Park, Joseph O’Brien, Carrie Jun Cai, Meredith Ringel Morris, Percy Liang, and Michael S. Bernstein. 2023. Generative agents: Interactive simulacra of human behavior. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology. find
  27. Sundar Pichai. 2024. Introducing Gemini 2.0: our new AI model for the agentic era. find
  28. Peng Qi, Yuyan Bu, Juan Cao, Wei Ji, Ruihao Shui, Junbin Xiao, Danding Wang, and Tat-Seng Chua. 2023a. Fakesv: A multimodal benchmark with rich social context for fake news detection on short video platforms. Proceedings of the AAAI Conference on Artificial Intelligence, 37(12):14444–14452. find
  29. Peng Qi, Yuyang Zhao, Yufeng Shen, Wei Ji, Juan Cao, and Tat-Seng Chua. 2023b. Two heads are better than one: Improving fake news video detection by correlating with neighbors. In Findings of the Association for Computational Linguistics: ACL 2023, pages 11947–11959. find
  30. Shengsheng Qian, Jinguang Wang, Jun Hu, Quan Fang, and Changsheng Xu. 2021. Hierarchical multi-modal contextual attention network for fake news detection. In Proceedings of the 44th International ACM SI-GIR Conference on Research and Development in Information Retrieval, page 153–162. find
  31. Lixiong Qin, Yang Zhang, Mei Wang, Jiani Hu, Weihong Deng, and Weiran Xu. 2025. Fake-in-facext: Towards fine-grained explainable deepfake analysis. Preprint, arXiv:2510.20531. — [arXiv:2510.20531](https://arxiv.org/abs/2510.20531) find
  32. Lanyu Shang, Ziyi Kou, Yang Zhang, and Dong Wang. 2021. A multimodal misinformation detector for covid-19 short videos on tiktok. In 2021 IEEE International Conference on Big Data (Big Data), pages 899–908. find
  33. Jing Shen, Yanjia Wang, Shengze Wang, Yuping Zhang, and Haibo Liu. 2025. Multi-modal similarity guided adaptive fusion network for short video fake news detection. In Proceedings of the 2025 International Conference on Multimedia Retrieval, ICMR 2025, Chicago, IL, USA, 30 June 2025 - 3 July 2025, pa… find
  34. Tomás Soucek and Jakub Lokoc. 2024. Transnet v2: An effective deep network architecture for fast shot transition detection. In Proceedings of the 32nd ACM International Conference on Multimedia, page 11218–11221. find
  35. Li Sun, Haoqi Zhang, Songyang Zhang, and Jiebo Luo. 2020. Content-based analysis of the cultural differences between tiktok and douyin. In 2020 IEEE International Conference on Big Data (Big Data), pages 4779–4786. find
  36. Sahar Tahmasebi, Eric Müller-Budack, and Ralph Ewerth. 2024. Multimodal misinformation detection using large vision-language models. In Proceedings of the 33rd ACM International Conference on Information and Knowledge Management, page 2189–2199. find
  37. Gemma Team, Aishwarya Kamath, Johan Ferret, Shreya Pathak, Nino Vieillard, Ramona Merhej, Sarah Perrin, Tatiana Matejovicova, Alexandre Ramé, Morgane Rivière, Louis Rouillard, Thomas Mesnard, Geoffrey Cideron, Jean bastien Grill, Sabela Ramos, Edouard Yvinec, Michelle Casbon, Etienne Pot, Ivo Penche… find
  38. Junxi Wang, Jize Liu, Na Zhang, and Yaxiong Wang. 2025a. Consistency-aware fake videos detection on short video platforms. In ICIC 2025, Ningbo, China, July 26-29, 2025, Proceedings, Part XVIII, pages 200–210. find
  39. JunXi Wang, Yaxiong Wang, Lechao Cheng, and Zhun Zhong. 2025b. FakeSV-VLM: Taming VLM for detecting fake short-video news via progressive mixture-of-experts adapter. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 4782–4798. find
  40. Yiting Wang. 2021. Multimodal analysis: Researching short-form videos and the theatrical practices. find
  41. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, brian ichter, Fei Xia, Ed Chi, Quoc V Le, and Denny Zhou. 2022. Chain-of-thought prompting elicits reasoning in large language models. In Advances in Neural Information Processing Systems, volume 35, pages 24824–24837.
  42. Jin Xu, Zhifang Guo, Jinzheng He, Hangrui Hu, Ting He, Shuai Bai, Keqin Chen, Jialin Wang, Yang Fan, Kai Dang, Bin Zhang, Xiong Wang, Yunfei Chu, and Junyang Lin. 2025a. Qwen2.5-omni technical report. Preprint, arXiv:2503.20215.
  43. Qingzheng Xu, Heming Du, Szymon Łukasik, Tianqing Zhu, Sen Wang, and Xin Yu. 2025b. Mdam3: A misinformation detection and analysis framework for multitype multimodal media. In Proceedings of the ACM on Web Conference 2025, page 5285–5296. find
  44. Hai Yu, Chong Deng, Qinglin Zhang, Jiaqing Liu, Qian Chen, and Wen Wang. 2025. Multimodal fusion and coherence modeling for video topic segmentation. In Findings of the Association for Computational Linguistics: ACL 2025, pages 17577–17593. find
  45. Liyuan Zhang, Yang Yajing, Yan Yang, Yong Liu, Zhongyan Gui, Ruofan Li, and Hao Fei. 2025a. MFSVFND: multimodal fusion network for detecting fake news on short video platforms. In ICMR 2025, Chicago, IL, USA, 30 June 2025 - 3 July 2025, pages 2123–2127. encoding, enabling it to process images of var… find
  46. Zongmeng Zhang, Wengang Zhou, Jie Zhao, and Houqiang Li. 2025b. Robust multimodal large language models against modality conflict. In Proceedings of the 42nd International Conference on Machine Learning. find
  47. InternVL3 (Zhu et al., 2025) introduces a native multimodal pre-training paradigm that unifies visual and linguistic learning from scratch. This model offers performance comparable to GPT-4o while maintaining strong pure language capabilities. We use it to detect short video news through Zero-Shot,… find
  48. Xiaofan Zheng, Zinan Zeng, Heng Wang, Yuyang Bai, Yuhan Liu, and Minnan Luo. 2025. From predictions to analyses: Rationale-augmented fake news detection with large vision-language models. In Proceedings of the ACM on Web Conference 2025, page 5364–5375. find
  49. Jianlong Zhou, Amir Gandomi, Fang Chen, and Andreas Holzinger. 2021. Evaluating the quality of machine learning explanations: A survey on methods and metrics. Electronics, 10:593. find
  50. Jinguo Zhu, Weiyun Wang, Zhe Chen, Zhaoyang Liu, Shenglong Ye, Lixin Gu, Hao Tian, Yuchen Duan, Weijie Su, Jie Shao, Zhangwei Gao, Erfei Cui, Xuehui Wang, Yue Cao, Yangzhou Liu, Xingguang Wei, Hongjie Zhang, Haomin Wang, Weiye Xu, and 32 others. 2025. Internvl3: Exploring advanced training and test-…

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/wang-2026-csi-investigative-multi-agent-framework/