成功大學博碩士論文系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	寒山 Niven, Timothy Stephen
論文名稱：	國家資助資訊操作之計算式分析與通報 Computational Analysis and Reporting of State-Sponsored Information Operations
指導教授：	李政德 Li, Cheng-Te
學位類別：	博士 Doctor
系所名稱：	電機資訊學院 - 資訊工程學系 Department of Computer Science and Information Engineering
論文出版年：	2026
畢業學年度：	114
語文別：	英文
論文頁數：	115
中文關鍵詞：	大型語言模型、機器學習分類、國家資助資訊操作、宣傳、敘事分析、知識本體
外文關鍵詞：	Large Language Models, Machine Learning Classification, State-sponsored Information Operations, Propaganda, Narrative Analysis, Ontologies
相關次數：	點閱：7 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

本論文對於國家資助資訊操作（state-sponsored information operations, SSIOs）的分析與通報提出三項關鍵貢獻。SSIOs 係指國家行為者為取得地緣政治優勢而試圖操縱資訊環境的行為。第一項貢獻在於診斷神經式自然語言處理（NLP）系統中的一項核心方法論陷阱：在推理任務上表現出的表面成功，往往是由淺層的統計線索所驅動，而非真正的推論能力。第二項貢獻為對 SSIO 分析的概念性推進：本文提出一個以敘事為中心的分析框架，將宣傳母題（propaganda tropes）置於核心分析單位，從而使對「參與式宣傳」這一高度困難問題的理解更加豐富且細緻。第三項貢獻為對 SSIO 防禦的實務性貢獻：本文提供首個具多位標註者、以事件層級為單位的 DISARM TTP 標註資料集，評估大型語言模型（LLM）的表現，並診斷與實際操作應用高度相關的失效模式。綜合而言，這些貢獻使以計算方法支援 SSIO 因應工作建立於嚴謹的方法論基礎之上，推動該領域超越對實務而言過於侷限或無關的概念框架，轉向更具生產力的理論架構，並即時為 SSIO 事件的通報工作提供實質協助，特別是在最具挑戰性的歸因問題上。

This thesis makes three key contributions to the analysis and reporting of state-sponsored information operations (SSIOs), defined as attempts by state actors to manipulate the information environment to gain a geopolitical advantage. The first contribution is to diagnose a core methodological pitfall in neural NLP systems: apparent success on reasoning tasks can be driven by superficial statistical cues rather than genuine inferential competence. The second contribution is a conceptual contribution to SSIO analysis: it advances a narrative-centered framework that foregrounds propaganda tropes as a core analytic unit, enabling richer understanding of the difficult problem of participatory propaganda. The third contribution is a practical contribution to SSIO defense: It delivers the first multi-annotator, incident-level dataset for DISARM TTP tagging, evaluates LLM performance, and diagnoses failure modes directly relevant to operational use. Taken together, these contributions put computational support to addressing SSIOs on a rigorous methodological footing, move the field beyond conceptual frameworks that are restrictive or irrelevant to the field of practice into productive conceptual frameworks, and immediately offer practical assistance in the reporting of SSIO incidents which helps with the most challenging problem of attribution.

摘要 i
Abstract ii
Acknowledgements iii
Table of Contents iv
List of Tables vii
List of Figures viii
Chapter 1. Introduction 1
1.1. Background and Motivation 1
1.2. Central Problem and Research Vision 2
1.3. Research Gaps 2
1.3.1. Methodological reliability 2
1.3.2. Over-narrow content framing 2
1.3.3. Practice-research disconnect 3
1.4. Overview of the Dissertation 3
1.5. Contributions of the Dissertation 4
1.5.1. Methodological Contribution: Evaluating Neural Network Classifiers for Robustness to Noise Distributions 4
1.5.2. Conceptual Contribution: Productively Analyzing the Content of SSIOs and Facilitating the Study of Participatory Propaganda 4
1.5.3. Contribution to the Field of Practice: Supporting Attribution and Reporting through TTP Tagging 4
Chapter 2. Background and Related Work 5
2.1. State-Sponsored Strategic Information Operations 5
2.2. Foreign Information Manipulation and Interference 5
2.3. Tactics, Techniques, and Procedures (TTPs) 5
2.4. DISARM 6
2.5. Computer-Assisted TTP Tagging 7
2.6. Propaganda Analysis 7
2.7. Participatory Propaganda 9
2.8. Narrative and Trope 10
2.9. Machine Learning Understanding of Argument 13
2.10. Exploitation of Spurious Statistics 14
Chapter 3. Methodology 15
3.1. Assessing Exploitation of Statistical Cues in the Argument Comprehension Reasoning Task 15
3.1.1. Task Description and Baselines 15
3.1.2. BERT 15
3.1.3. Statistical Cues 16
3.1.4. ARCT Probing Experiments 17
3.1.5. Adversarial ARCT Test Set 17
3.2. Computational Analysis of Propaganda Tropes and Participatory Propaganda 18
3.2.1. Data Collection 18
3.2.2. Codebook Development and Annotation 19
3.2.3. Annotation 21
3.2.4. Machine Learning 22
3.2.5. Engagement Metric Comparisons 24
3.3. Automatic Tagging of DISARM TTPs 24
3.3.1. Dataset Construction 24
3.3.2. Detection Model 29
Chapter 4. Results 33
4.1. BERT Learns Exclusively from ARCT’s Noise Distribution 33
4.1.1. Baseline and BERT Results 33
4.1.2. Analysis of Spurious Statistics in the ARCT Dataset 33
4.1.3. Probing Experiments 34
4.1.4. Adversarial Test Set 34
4.1.5. Discussion 35
4.2. Propaganda Tropes Facilitate SSIO and Participatory Propaganda Research 36
4.2.1. RQ1: Human Identification of Tropes 36
4.2.2. RQ2: Machine Learning Identification of Tropes 37
4.2.3. RQ3: Prevalence of Tropes 40
4.2.4. RQ4: Trope Engagement Metrics 41
4.3. Limitations 45
4.3.1. Discussion 47
4.4. LLMs Cannot Automate DISARM Tagging but are Useful Assistants 47
4.4.1. RQ1: How challenging is this task? 48
4.4.2. RQ2: What is the performance-to-cost trade-off? 50
4.4.3. RQ3: Does supplying additional annotation guidelines help? 50
4.4.4. RQ4: How are LLMs failing? 51
4.4.5. Limitations 53
4.4.6. Discussion 53
Chapter 5. Conclusion 54
References 56
Appendix A. Propaganda Tropes 81
A.1. Codebook 81
A.1.1. Amazing Xinjiang 81
A.1.2. Happy Uyghurs 82
A.1.3. Law and Order 82
A.1.4. Magnificent Development 83
A.1.5. Powerful Ancestral Motherland 83
A.1.6. Perfect Hong Kong Democracy 83
A.1.7. Red Hong Kong 83
A.1.8. Social Progress 84
A.1.9. Vandals and Traitors 84
A.1.10. Western Hegemony 84
A.2. Detailed Machine Learning Results 84
A.2.1. Machine Learning Ablation Results 84
A.2.2. Cross-Validation Results 87
A.2.3. Co-Occurrence of Text and Image Labels 87
A.3. Engagement Metric Comparisons 91
A.4. Inter-Annotator Agreement Supplement 95
A.4.1. Per-Trope Agreement 95
A.4.2. Multi-Label Statistics 95
A.4.3. Per-Trope Confusion Matrices 95

Appendix B. DISARM Tagging 97
B.1. Coding Guidance 97
B.2. Prompt 100
B.3. Error Analysis 102
                                    

[1] Adam. DISARM Red Retrospective and Futrospective. Medium, Aug. 2025.
[2] F. Alam, S. Cresci, T. Chakraborty, F. Silvestri, D. Dimitrov, G. D. S. Da San Martino, S. Shaar, H. Firooz, and P. Nakov. A survey on multimodal disinformation detection. In Proceedings of the 29th International Conference on Computational Linguistics, pages 6625–6643, Republic of Korea, 2022.
[3] Md Tanvirul Alam, Dipkamal Bhusal, Youngja Park, and Nidhi Rastogi. Looking Beyond IOCs: Automatically Extracting Attack Patterns from External CTI. In Proceedings of the 26th International Symposium on Research in Attacks, Intrusions and Defenses, RAID. ACM, 2023.
[4] B. Alpermann and M. Malzer. “in other news”: China’s international media strategy on xinjiang—cgtn and new china tv on youtube. Modern China, 0(0).
[5] A. Alvarez and J. Font. Tropetwist: Trope-based narrative structure generation. In Proceedings of the 17th International Conference on the Foundations of Digital Games, pages 1–8, New York, USA, 2022. Association for Computing Machinery.
[6] A Badawy, Emilio Ferrara, and Kristina Lerman. Analyzing the digital traces of political manipulation: The 2016 Russian interference Twitter campaign. In 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), pages 258–265. IEEE, 2018.
[7] A. Barrón-Cedeño, I. Jaradat, G. Da San Martino, and P. Nakov. Proppy: Organizing the news based on their propagandistic content. Information Processing & Management, 56(5):1849–1864, 2019.
[8] R. Barthes. An introduction to the structural analysis of narrative. New Literary History, 6(2):237–272, 1975.
[9] M. Bengtsson. How to plan and perform a qualitative study using content analysis. NursingPlus Open, 2:8–14, 2016.
[10] A. Biswas, T. Niven, and Y.-R. Lin. The dynamics of political narratives during the russian invasion of ukraine. In Social, Cultural, and Behavioral Modeling, pages 33–43, Cham, 2023. Springer Nature Switzerland.
[11] L. Bittman. The KGB and Soviet disinformation: An insider’s view. Pergamon-Brassey's, 1985.
[12] Elizabeth Black and Anthony Hunter. A relevance-theoretic framework for constructing and deconstructing enthymemes. J. Log. Comput., 22:55–78, 02 2012.
[13] Teresa Botschen, Daniil Sorokin, and Iryna Gurevych. Frame- and entity-based knowledge for common-sense argumentative reasoning. In ArgMining@EMNLP, 2018.
[14] Samuel R. Bowman, Gabor Angeli, Christopher Potts, and Christopher D. Manning. A large annotated corpus for learning natural language inference. CoRR, abs/1508.05326, 2015.
[15] A.-M. Brady. Marketing Dictatorship: Propaganda and Thought Work in Contemporary China. Rowman & Littlefield, UK, 1 edition, 2008.
[16] Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, et al. Language models are few-shot learners. Advances in Neural Information Processing Systems, 33:1877–1901, 2020.
[17] Marvin Büchel, Tommaso Paladini, Stefano Longari, Michele Carminati, Stefano Zanero, Hodaya Binyamini, Gal Engelberg, Dan Klein, Giancarlo Guizzardi, Marco Caselli, Andrea Continella, Maarten van Steen, Andreas Peter, and Thijs van Ede. SoK: Automated TTP Extraction from CTI Reports – Are We There Yet? In Proceedings of the 34th USENIX Security Symposium, 2025.
[18] R. L. Bytwerk. Grassroots propaganda in the third reich: The reich ring for national socialist propaganda and public enlightenment. German Studies Review, 33(1):93–118, 2010.
[19] Alina Bârgăoanu and Mihaela Pană. Cyber Influence Defense: Applying the DISARM Framework to a Cognitive Hacking Case from the Romanian Digital Space. Applied Cybersecurity & Internet Governance, 3(1):91–121, 2024.
[20] G. Cánovas López de Molina, F. Sánchez González, P. Nespoli, J. Pastor Galindo, and J.A. Ruipérez Valiente. Analyzing frameworks to model disinformation attacks in online social networks. In Jornadas Nacionales de Investigación en Ciberseguridad (JNIC), pages 92–99, Sevilla, 2024. Universidad de Sevilla. Escuela Técnica Superior de Ingeniería Informática.
[21] M. Caulfield. Tropes and networked digital activism 1: Trope-field fit. Online, June 2021.
[22] Danilo Cavaliere, Giuseppe Fenza, Domenico Furno, and Vincenzo Loia. A semantic model bridging DISARM framework and Situation Awareness for disinformation Attacks Attribution. In 2024 IEEE Conference on Cognitive and Computational Aspects of Situation Management (CogSIMA), pages 55–62, 2024.
[23] Qian Chen, Xiaodan Zhu, Zhen-Hua Ling, Si Wei, and Hui Jiang. Enhancing and combining sequential and tree LSTM for natural language inference. CoRR, abs/1609.06038, 2016.
[24] Yutong Cheng, Osama Bajaber, Saimon Amanuel Tsegai, Dawn Song, and Peng Gao. CTINEXUS: Leveraging Optimized LLM In-Context Learning for Constructing Cybersecurity Knowledge Graphs under Data Scarcity. CoRR, abs/2410.21060, 2024.
[25] China News Service. 就美方以涉嫌強迫勞動為由對若干中國企業采取限制措施 …. Twitter/X, September 2020. “In response to the U.S.’s restrictive measures against several Chinese companies on the grounds of suspected forced labor”.
[26] China News Service. 新疆是個好地方. Twitter/X, September 2022. “Xinjiang is a good place”.
[27] HongSeok Choi and Hyunju Lee. Gist at semeval-2018 task 12: A network transferring inference knowledge to argument reasoning comprehension task. In Proceedings of The 12th International Workshop on Semantic Evaluation, pages 773–777. Association for Computational Linguistics, 2018.
[28] Alexis Conneau, Douwe Kiela, Holger Schwenk, Loïc Barrault, and Antoine Bordes. Supervised learning of universal sentence representations from natural language inference data. CoRR, abs/1705.02364, 2017.
[29] Domenico Cotroneo, Roberto Natella, and Vittorio Orbinato. Elevating Cyber Threat Intelligence Against Disinformation Campaigns with Llm-Based Concept Extraction and the Fakecti Dataset. SSRN Electronic Journal, 2024.
[30] Fiona Coulston, Fiona Lynch, and Danya F. Vears. Collaborative coding in inductive content analysis: Why, when, and how to do it. J Genet Couns, 34(3):e70030, Jun. 2025.
[31] R. Cover and C. Milne. The “bury your gays” trope in contemporary television: Generational shifts in production responses to audience dissent. The Journal of Popular Culture, 56:810–823, 2023.
[32] E. D. Cubuk, B. Zoph, J. Shlens, and Q. Le. Randaugment: Practical automated data augmentation with a reduced search space. In Advances in Neural Information Processing Systems, volume 33, pages 18613–18624, 2020.
[33] Y. Cui, W. Che, T. Liu, B. Qin, and Z. Yang. Pre-training with whole word masking for chinese bert. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 29:3504–3514, 2021.
[34] G. Da San Martino, S. Yu, A. Barrón-Cedeño, R. Petrov, and P. Nakov. Finegrained analysis of propaganda in news article. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 5636–5646, 2019.
[35] 人民日報 (People’s Daily). 【新疆哈密：戈壁建出“太陽城＂】近年來，新疆哈密市依托當地豐富太陽能資源不斷加大太陽能發電項目的建設力度，目前已有 20 多家企業落戶哈密，建成有熔鹽塔式光熱發電電站，單晶矽電板、多晶矽電板、聚光光伏電板、薄膜光伏電板等光伏發電電站，裝機規模突破 170 萬千瓦。([hami, xinjiang: ”solar city” built in the gobi desert] in recent years, hami city, xinjiang has continuously increased the construction of solar power generation projects relying on the rich local solar energy resources. currently, more than 20 companies have settled in hami, and built molten salt tower solar thermal power stations, single crystal silicon panels, polycrystalline silicon panels, concentrated photovoltaic panels, thin-film photovoltaic panels and other photovoltaic power stations, with an installed capacity of over 1.7 million kilowatts.), December 2020.
[36] 人民日報 (People’s Daily). 【涉疆問題發布會戳穿西方謊言：對潑向新疆的污泥濁水、將堅定堅決懟回去！】新疆維吾爾自治區人民政府新聞辦公室 21 日在北京外交部舉行第二十一場涉疆問題新聞發布會，自治區政府官員和普通民眾回答了外國媒體記者提出的一系列尖銳問題 ([the press conference on xinjiangrelated issues exposed western lies: We will firmly and resolutely fight back against the mud and dirty water thrown at xinjiang! the information oﬀice of the people’s government of the xinjiang uygur autonomous region held the 21st press conference on xinjiang-related issues at the ministry of foreign affairs in beijing on the 21st. autonomous region government oﬀicials and ordinary people answered a series of sharp questions raised by foreign media reporters.), December 2020.
[37] 人民日報 (People’s Daily). 【# 世界首條沙漠環線鐵路正式閉合】塔克拉瑪干沙漠是世界第二大流動性沙漠，也是中國最大的沙漠。今天，環塔克拉瑪干沙漠鐵路在新疆和田民豐縣境內完成最後的鋪軌作業，這意味著世界上首條沙漠環線鐵路正式閉合。和若鐵路修好形成環線之後，向東經過若羌就可以直接出疆，減少了一千多公里的行程。([#the world’s first desert loop railway oﬀicially closed] the taklimakan desert is the second largest mobile desert in the world and the largest desert in china. today, the taklimakan desert railway completed the final track laying operation in minfeng county, hotan, xinjiang, which means that the world’s first desert loop railway has oﬀicially closed. after the hotan-ruoqiang railway is completed and forms a loop line, passengers can directly leave xinjiang via ruoqiang to the east, reducing the journey by more than a thousand kilometers.), September 2021.
[38] 人民日報 (People’s Daily). 【北斗“護航＂新疆百萬畝棉花開播】目前新疆從南到北的棉花春播已經開始，3 月 26 日，裝有北斗衛星定位導航系統的播種機在新疆生產建設兵團第三師四十九團的農田裡播種棉花，該師今年計劃播種棉花 100 餘萬畝。隨著天氣轉暖，喀什地區莎車縣、兵團第三師圖木舒克市等地的棉花全面開播。([beidou ”escorts” xinjiang’s one million acres of cotton planting] currently, spring cotton planting has begun from south to north in xinjiang. on march 26, a seed drill equipped with the beidou satellite positioning and navigation system was sowing cotton in the farmland of the 49th regiment of the third division of the xinjiang production and construction corps. the division plans to sow more than 1 million acres of cotton this year. as the weather warms up, cotton planting has started in full swing in shache county, kashgar prefecture, tumushuke city, the third division of the xinjiang production and construction corps, and other places.), March 2021.
[39] 人民日報 (People’s Daily). 【南京至蒂爾堡中歐班列首發開行】5 月 10 日，在南京堯化門貨場，江蘇南京至荷蘭蒂爾堡中歐班列從南京堯化門貨場首發開行。班列滿載液晶顯示屏、工業機械配件、紡織服飾等 100 個標箱貨物，將經由新疆阿拉山口出境，途經哈薩克斯坦、俄羅斯、白俄羅斯等國後抵達荷蘭蒂爾堡市，全程用時 18 天左右。([the first china-europe freight train from nanjing to tilburg started running] on may 10, at nanjing yaohuamen freight yard, the china-europe freight train from nanjing, jiangsu to tilburg, the netherlands started running for the first time. the train is loaded with 100 standard containers of goods including lcd screens, industrial machinery parts, textiles and clothing. it will leave the country through alashankou in xinjiang, and arrive in tilburg, the netherlands after passing through kazakhstan, russia, belarus and other countries. the whole journey will take about 18 days.), May 2021.
[40] 人民日報 (People’s Daily). 【帥氣！直擊零下 30 攝氏度射擊訓練】近日，新疆軍區某團將部隊拉至野外展開訓練，在零下 30 攝氏度的嚴寒條件下，開展車載重機槍射擊、步兵多種姿勢射擊、野戰自炊保障等課目訓練。（圖源：解放軍報）, 2021.
[41] 人民日報 (People’s Daily). 【摘掉有色眼鏡才能看見這抹白】今天的《新疆日報》提及，此前新疆曾多次邀請歐盟駐華使節參訪，但他們以各種理由百般拖延，甚至提出無理要求。說白了就是不敢面對真正的新疆，害怕新疆社會穩定發展、人民安居樂業的繁榮景象戳穿他們的謊言。西方那些戴有色眼鏡、選擇性失明的人，才看不見這抹白。([you can only see this whiteness when you take off your tinted glasses] today’s xinjiang daily mentioned that xinjiang had invited eu ambassadors to china to visit many times before, but they delayed for various reasons and even made unreasonable demands. to put it bluntly, they are afraid to face the real xinjiang and are afraid that the prosperous scene of xinjiang’s social stability and development, with people living and working in peace and contentment, will expose their lies. only those people in the west who wear tinted glasses and are selectively blind cannot see this whiteness.), March 2021.
[42] 人民日報 (People’s Daily). 【經民聯希望推動香港回歸經濟城市定位，樹立 “發展惠民＂主調】香港經濟民生聯盟（經民聯）6 月 18 日舉行記者會發布“七一宣言＂，提出在新時代的兩大口號“推經濟＂和“惠民生＂，並圍繞上述口號提出四大綱領，希望推動香港回歸經濟城市的定位，重新樹立“發展惠民＂主調。, 2021.
[43] 人民日報 (People’s Daily). 【被玩弄的民主騙局再不能蒙蔽世人和港人】國務院港澳事務辦公室發言人 24 日發表談話表示，美國等少數西方國家一再惺惺作態對香港民主發展“表示關切＂，但其真實目的是要搞亂香港以遏制中國，被他們玩弄的民主騙局如此虛偽險惡，實在可鄙！([the democratic scam that has been played can no longer deceive the world and hong kong people] a spokesperson for the hong kong and macao affairs oﬀice of the state council made a statement on the 24th, saying that the united states and a few other western countries have repeatedly pretended to ”express concern” about the development of democracy in hong kong, but their real purpose is to mess up hong kong in order to contain china. the democratic scam they have played is so hypocritical and sinister, it is really despicable!), December 2021.
[44] 人民日報 (People’s Daily). 【風雪中，升級版極寒軌道除雪車出動了！】進入冬季，位於新疆克塔鐵路區間經常出現風吹雪現象，容易導致鐵路線積雪板結，影響列車運行。對此，該路段引進升級版極寒軌道除雪車，這種除雪車能在零下 40 攝氏度工作，集高速掃雪、碎板結、拋雪、推鏟雪、吹雪和軌道牽引功能於一體，有效保障鐵路暢通。([in the snow and wind, the upgraded extreme cold track snowplow was dispatched!] as winter approaches, wind-blown snow often occurs in the keta railway section in xinjiang, which can easily cause snow to accumulate on the railway line and affect train operations. in response to this, the section introduced an upgraded extreme cold rail snowplow, which can work at minus 40 degrees celsius. it integrates the functions of high-speed snow sweeping, breaking up compacted snow, throwing snow, pushing and shoveling snow, blowing snow and track traction, effectively ensuring the smooth operation of the railway.), December 2021.
[45] 人民日報 (People’s Daily). 第二十四屆北京• 香港經濟合作研討洽談會 6 日以線上方式，在北京、香港兩地同時開幕，吸引 500 余名相關專業人士在北京和香港會場參會。香港特區行政長官林鄭月娥表示，北京與香港的合作源遠流長，根基深厚，期待未來兩地在高等教育和創新科技、金融、文化藝術三方面深化合作。(the 24th beijing-hong kong economic cooperation seminar and negotiation conference opened online simultaneously in beijing and hong kong on the 6th, attracting more than 500 relevant professionals to attend the conference at the venues in beijing and hong kong. the chief executive of the hong kong special administrative region, carrie lam, said that the cooperation between beijing and hong kong has a long history and deep roots, and she looks forward to deepening cooperation between the two places in the three areas of higher education and innovative technology, finance, and culture and art in the future.), December 2021.
[46] 人民日報 (People’s Daily). 【新疆首個高高原機場正式試飛】18 日，一架空客 a319-115 型客機平穩降落跑道，標誌著新疆首個高高原機場——塔什庫爾干紅其拉甫機場正式開啟實地試飛驗證工作。中國民航規定，海拔高度在 2438 米及以上的機場為高高原機場。該機場標高 3258.4 米，年旅客吞吐量可達 16 萬人次、貨郵吞吐量達 400 噸。([xinjiang’s first high-plateau airport oﬀicially starts test flight] on the 18th, an airbus a319-115 passenger plane landed smoothly on the runway, marking the oﬀicial start of field test flight verification work at xinjiang’s first high-plateau airport, taxkorgan hongqilapu airport. according to the china civil aviation administration, airports with an altitude of 2,438 meters or above are high-plateau airports. the airport is located at an elevation of 3,258.4 meters, with an annual passenger throughput of 160,000 and a cargo and mail throughput of 400 tons.), May 2022.
[47] S. Dash, A. Arya, S. Kaur, and J. Pal. Narrative building in propaganda networks on indian twitter. In Proceedings of the 14th ACM Web Science Conference, pages 239–244, 2022.
[48] Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. BERT: pre-training of deep bidirectional transformers for language understanding. CoRR, abs/1810.04805, 2018.
[49] D. Dimitrov, B. Bin Ali, S. Shaar, F. Alam, F. Silvestri, H. Firooz, P. Nakov, and G. Da San Martino. Detecting propaganda techniques in memes. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 6603–6617, 2021.
[50] Renee DiResta. The new media goliaths, 2023.
[51] DISARM Foundation. DISARM Navigator. Web application.
[52] DISARM Foundation. DISARMframeworks: Master copies of the DISARM frameworks, with generated files to help you explore the data, 2025.
[53] Doublethink Lab. The Spread of Ukraine Biolabs Conspiracy Content in Chinese on Twitter. Medium, Sep. 2022.
[54] Doublethink Lab. Artificial Multiverse: Foreign Information Manipulation and Interference in Taiwan＇s 2024 National Elections. Medium, Aug. 2024.
[55] Doublethink Lab. Doublethink Lab and Voice of America: US Election Joint Observation Mission Final Report. Medium, Dec. 2024.
[56] Doublethink Lab. Watch Our Neighbourhood: Collaborative Approaches to Address FIMI During the India Election. Medium, Nov. 2024.
[57] Doublethink Lab. Weaponizing the Sinovac Slander: Online CIB Against the United States and the Philippine President. Medium, Dec. 2024.
[58] Doublethink Lab. Inauthentic Accounts Impersonate Taiwanese to Attack Political Party. Medium, Jul. 2025.
[59] European Commission. Ukraine: Sanctions on Kremlin-backed outlets Russia Today and Sputnik, Mar. 2022.
[60] European External Action Service. 1st EEAS Report on Foreign Information Manipulation and Interference (FIMI) Threats. Technical report, European External Action Service, 2023. Accessed: 2025-08-31.
[61] European External Action Service. 3rd EEAS Report on Foreign Information Manipulation and Interference Threats, Mar. 2025.
[62] European External Action Service. Information integrity and countering foreign information manipulation & interference (fimi). European Union, 2025.
[63] European External Action Service. Information Integrity and Countering Foreign Information Manipulation & Interference (FIMI), 2025.
[64] L. Frermann, J. Li, S. Khanehzar, and G. Mikolajczak. Conflicts, villains, resolutions: Towards models of narrative media framing. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8712–8732, 2023.
[65] Margot Fulde-Hardy. Foreign Information Manipulation and Interference (FIMI) Activities Targeting Elections (2014-2024): Presenting a dataset, a methodology, and a codebook to guide future applications of structured frameworks enabling threat assessment. Working paper, Columbia University, School of International and Public Affairs, 2024.
[66] D. Gala, M. O. Khursheed, H. Lerner, B. O’Connor, and M. Iyyer. Analyzing gender bias within narrative tropes. In Proceedings of the Fourth Workshop on Natural Language Processing and Computational Social Science, pages 212–217, 2020.
[67] A. Ganti, E. A. H. Hussein, S. Wilson, Z. Ma, and X. Zhao. Narrative style and the spread of health misinformation on twitter. In Findings of the Association for Computational Linguistics: EMNLP, pages 4266–4282, 2023.
[68] P. García-Sánchez, A. Velez-Estevez, J. Julián Merelo, and M. J. Cobo. The simpsons did it: Exploring the film trope space and its large scale structure. PLOS ONE, 16(3), 2021.
[69] G. Gérard. Narrative Discourse: An Essay in Method. Cornell University Press, 1980.
[70] Nathaniel Gleicher. Coordinated inauthentic behavior explained, 2018.
[71] Max Glockner, Vered Shwartz, and Yoav Goldberg. Breaking NLI systems with sentences that require simple lexical inferences. CoRR, abs/1805.02266, 2018.
[72] Felipe Sánchez González, Javier Pastor-Galindo, and José A. Ruipérez-Valiente. DISINFOX: an open-source threat exchange platform serving intelligence on disinformation and influence operations, 2025.
[73] A. Y. Guevara. Propaganda in mexico’s drug war. Journal of Strategic Security, 6(3):131–151, 2013.
[74] X. Guo and S. Vosoughi. Multi-modal identification of state-sponsored propaganda on social media. In 2020 25th International Conference on Pattern Recognition, pages 10576–10583, 2021.
[75] Suchin Gururangan, Swabha Swayamdipta, Omer Levy, Roy Schwartz, Samuel R. Bowman, and Noah A. Smith. Annotation artifacts in natural language inference data. CoRR, abs/1803.02324, 2018.
[76] K. Gwet. Kappa statistic is not satisfactory for assessing the extent of agreement between raters, 2002.
[77] Ivan Habernal, Judith Eckle-Kohler, and Iryna Gurevych. Argumentation mining on the web from information seeking perspective. In ArgNLP, 2014.
[78] Ivan Habernal, Henning Wachsmuth, Iryna Gurevych, and Benno Stein. The argument reasoning comprehension task: Identification and reconstruction of implicit warrants. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), pages 1930–1940, New Orleans, Louisiana, June 2018. Association for Computational Linguistics.
[79] Ivan Habernal, Henning Wachsmuth, Iryna Gurevych, and Benno Stein. Semeval2018 task 12: The argument reasoning comprehension task. In Proceedings of The 12th International Workshop on Semantic Evaluation, pages 763–772. Association for Computational Linguistics, 2018.
[80] Jacob Harrison. LangChain: Building applications with LLMs through composability. https://github.com/langchain-ai/langchain, 2022.
[81] Sepp Hochreiter and Jürgen Schmidhuber. Long short-term memory. Neural computation, 9(8):1735–1780, 1997.
[82] H. Hoeken, P. Swanepoel, E. Saal, and C. Jansen. Using message form to stimulate conversations: The case of tropes. Communication Theory, 19(1):49–65, 2009.
[83] L. Hu. A two-step method for classifying political partisanship using deep learning models. Social Science Computer Review, 0(0), 2023.
[84] B. Hughes, C. Miller-Idriss, R. Piltch-Loeb, K. White, M. Creizis, C. Cain, and E. Savoia. Development of a codebook of online anti-vaccination rhetoric to manage covid-19 vaccine misinformation. medRxiv, 2021.
[85] Ghaith Husari, Xi Niu, Bill Chu, and Ehab Al-Shaer. Using Entropy and Mutual Information to Extract Threat Actions from Cyber Threat Intelligence. In 2018 IEEE International Conference on Intelligence and Security Informatics, ISI. IEEE, 2018.
[86] Eric M. Hutchins, Michael J. Cloppert, and Rohan M. Amin. Intelligence-Driven Computer Network Defense Informed by Analysis of Adversary Campaigns and Intrusion Kill Chains. In J. R. C. Nurse, editor, Leading Issues in Information Warfare and Security Research, volume 1, pages 80–106. Academic Publishing International, 2011.
[87] C. Jack. Lexicon of lies: Terms for problematic information. Data & Society, 2017.
[88] Robin Jia and Percy Liang. Adversarial examples for evaluating reading comprehension systems. CoRR, abs/1707.07328, 2017.
[89] Jason Jo and Yoshua Bengio. Measuring the tendency of cnns to learn surface statistical regularities. CoRR, abs/1711.11561, 2017.
[90] Piotr Marszałek Kamil Baraniuk. The potential of Cyber Threat Intelligence analytical frameworks in research on information operations and influence operations. Internal Security Review, 2024(Issue 31 (16)):279–320, 2024.
[91] G. H. Karlsen. Divide and rule: Ten lessons about russian political influence activities in europe. Palgrave Communications, 5(19), 2019.
[92] A. Kazemi, K. Garimella, D. Gaffney, and S. Hale. Claim matching beyond english to scale global fact-checking. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 4504–4517, 2021.
[93] Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization. CoRR, abs/1412.6980, 2014.
[94] J. R. Landis and G. G. Koch. The measurement of observer agreement for categorical data. Biometrics, 33(1):159–174, 1977.
[95] Valentine Legoy, Marco Caselli, Christin Seifert, and Andreas Peter. Automated Retrieval of ATT&CK Tactics and Techniques for Cyber Threat Reports. CoRR, abs/2004.14322, 2020.
[96] E. Levi, G. Mor, T. Sheafer, and S. Shenhav. Detecting narrative elements in informational text. In Findings of the Association for Computational Linguistics: NAACL, pages 1755–1765, 2022.
[97] Stephan Lewandowsky. Fake news and participatory propaganda. In Cognitive Illusions. Routledge, 3 edition, 2022.
[98] Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. Retrieval-augmented generation for knowledge-intensive NLP tasks. In Proceedings of the 34th Conference on Neural Information Processing Systems (NeurIPS 2020), pages 9459–9474, 2020.
[99] Zhenyuan Li, Jun Zeng, Yan Chen, and Zhenkai Liang. AttackG: Constructing Technique Knowledge Graph from Cyber Threat Intelligence Reports. In Computer Security - ESORICS 2022 - 27th European Symposium on Research in Computer Security, Lecture Notes in Computer Science. Springer, 2022.
[100] Marco Lippi and Paolo Torroni. Argumentation mining: State of the art and emerging trends. ACM Trans. Internet Technol., 16(2):10:1–10:25, March 2016.
[101] A. Loftus. Eu investigates tiktok over alleged russian meddling in romanian vote. BBC News, 2024.
[102] A. Loftus. EU investigates TikTok over alleged Russian meddling in Romanian vote. BBC News, 2024.
[103] L. R. Luqiu. Constructing its own reality: The ccp’s agenda for the hong kong anti-extradition bill movement. In Research Handbook on Political Propaganda, pages 143–155. Edward Elgar Publishing, UK, 2021.
[104] Wenhao Ma. China-connected spamouflage impersonated Dutch cartoonist. Voice of America, Sep. 2024.
[105] A. Maarouf, D. Bär, D. Geissler, and S. Feuerriegel. Hqp: A human-annotated dataset for detecting online propaganda. arXiv, 2023.
[106] H. K. Mak, K. K. Yau, and B. P. Chan. Prevalence-adjusted bias-adjusted kappa values as additional indicators to measure observer agreement. Radiology, 232(1):302–303, 2004.
[107] H. B. Mann and D. R. Whitney. On a test of whether one of two random variables is stochastically larger than the other. The Annals of Mathematical Statistics, 18(1):50–60, 1947.
[108] R. Thomas McCoy, Ellie Pavlick, and Tal Linzen. Right for the wrong reasons: Diagnosing syntactic heuristics in natural language inference. CoRR, abs/1902.01007, 2019.
[109] 观察者网新媒体 (Observer New Media). 【美媒曝 5 名乱港暴徒潜逃经过：想留台湾被拒，美国务院介入后赴美】香港“修例风波＂期间，一批乱港暴徒畏罪潜逃，其中 5 人被曝于 2020 年 7 月偷渡台湾，先抵达东沙群岛，后被秘密安置在高雄，当时，台当局对此事三缄其口。([us media revealed how five hong kong rioters escaped: they wanted to stay in taiwan but were rejected, and went to the us after the us state department intervened] during the hong kong ”amendment storm”, a group of hong kong rioters fled for fear of crime. five of them were exposed to have smuggled into taiwan in july 2020, first arriving at the dongsha islands, and then secretly settled in kaohsiung. at that time, the taiwan authorities remained silent on the matter.), June 2021.
[110] L. Miao, M. Last, and M. Litvak. Detecting troll tweets in a bilingual corpus. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 6247–6254, 2020.
[111] Marco Minici, Luca Luceri, Francesco Fabbri, and Emilio Ferrara. IOHunter: Graph Foundation Model to Uncover Online Information Operations. Proceedings of the AAAI Conference on Artificial Intelligence, 39(27):28258–28266, Apr. 2025.
[112] Massimo Minici, Francesco Cinus, Leonardo Luceri, and Emilio Ferrara. Uncovering coordinated cross-platform information operations: Threatening the integrity of the 2024 U.S. presidential election. First Monday, 29(11), 2024.
[113] MITRE Corporation. MITRE ATT&CK Framework. https://attack.mitre.org/. Accessed: 2025-08-23.
[114] Raquel Mochales and Marie-Francine Moens. Argumentation mining. Artif. Intell. Law, 19(1):1–22, March 2011.
[115] N. Nasrin, K.-K. R. Choo, M. Ko, and A. Rios. How many users are enough? exploring semi-supervision and stylometric features to uncover a russian troll farm. In Proceedings of the Second Workshop on Natural Language Processing for Internet Freedom: Censorship, Disinformation, and Propaganda, pages 20–30, 2019.
[116] R. A. Nelson. A Chronology and Glossary of Propaganda in the United States. Greenwood Press, 1996.
[117] David Nemer. Disentangling brazil＇ s disinformation insurgency. NACLA Report on the Americas, 53(4):406–413, 2021.
[118] China Global Television Network. Amazing xinjiang, Oct 2020.
[119] B. Nimmo, M. Torrey, M. Franklin, D. Agranovich, M. Milam, L. Hundley, and R. Flaim. Meta quarterly adversarial threat report q2 2023. Meta, August 2023.
[120] T. S. Niven. The spread of ukraine biolabs conspiracy content in chinese on twitter. Doublethink Lab, September 2022.
[121] T. S. Niven. The evolution of china’s interference in taiwan. The Diplomat, December 2023.
[122] Timothy Niven and Maiko Ichihara. To Influence Japan, China Tries Subtlety. American Purpose, Oct. 2021.
[123] Timothy Niven and Hung-Yu Kao. NLITrans at SemEval-2018 task 12: Transfer of semantic knowledge for argument comprehension. In Proceedings of The 12th International Workshop on Semantic Evaluation, pages 1099–1103, New Orleans, Louisiana, June 2018. Association for Computational Linguistics.
[124] L. S. Nowell, J. M. Norris, D. E. White, and N. J. Moules. Thematic analysis: Striving to meet the trustworthiness criteria. International Journal of Qualitative Methods, 16(1), 2017.
[125] W. M. O’Barr. Public service advertising and propaganda. Advertising & Society Review, 13(2), 2012.
[126] 观察者网 (Observer). 【华春莹在记者会上展示视频：美方导演的这出戏，该落幕了】当年参与策动伊拉克战争的美国前高官亲口承认，所谓新疆维吾尔族问题，只不过是美国企图从内部搞乱中国、遏制中国的战略阴谋 ([hua chunying showed a video at the press conference: This drama directed by the us should end] a former senior us oﬀicial who participated in instigating the iraq war admitted that the so-called xinjiang uyghur issue is nothing more than a strategic conspiracy by the united states to disrupt china from within and contain china.), March 2021.
[127] 观察者网 (Observer). 【# 新疆人 v.s. 外媒镜头下的“新疆人＂】([#xinjiang people v.s. ”xinjiang people” in the lens of foreign media]), February 2021.
[128] The Commisioner’s Oﬀice of China’s Foreign Ministry in the Hong Kong S.A.R. Hong kong will surely embrace prosperity and sound governance and those who go against the trend have no future, July 2022.
[129] Embassy of the People’s Republic of China in Angola. Us hegemony and its perils, February 2023.
[130] V.-A. Oliinyk, V. Vysotska, Y. Burov, K. Mykich, and V. Basto-Fernandes. Propaganda detection in text data based on nlp and machine learning. In Proceedings of the 2nd International Workshop on Modern Machine Learning Technologies and Data Science, pages 132–144, 2020.
[131] Jonathan Ong and Jason Vincent Cabañes. Architects of networked disinformation, 2018.
[132] C. Y. Park, J. Mendelsohn, A. Field, and Y. Tsvetkov. Challenges and opportunities in information manipulation detection: An examination of wartime russian media. In Findings of the Association for Computational Linguistics: EMNLP, pages 5209–5235, 2022.
[133] Javier Pastor-Galindo, Pantaleone Nespoli, José A. Ruipérez-Valiente, and David Camacho. Influence Operations in Social Networks, 2025.
[134] F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay. Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research, 12:2825–2830, 2011.
[135] Jeffrey Pennington, Richard Socher, and Christopher D. Manning. Glove: Global vectors for word representation. In Empirical Methods in Natural Language Processing (EMNLP), pages 1532–1543, 2014.
[136] J. Piskorski, N. Stefanovitch, G. Da San Martino, and P. Nakov. Semeval-2023 task 3: Detecting the category, the framing, and the persuasion techniques in online news in a multi-lingual setup. In Proceedings of the 17th International Workshop on Semantic Evaluation, pages 2343–2361, 2023.
[137] China Plus. Amazing xinjiang, April 2021.
[138] Adam Poliak, Jason Naradowsky, Aparajita Haldar, Rachel Rudinger, and Benjamin Van Durme. Hypothesis only baselines in natural language inference. CoRR, abs/1805.01042, 2018.
[139] F. Polletta and B. Gardner. Narrative and social movements. In Oxford Handbook of Social Movements, pages 534–538. Oxford University Press, New York, 2015.
[140] Francesca Polletta and Jessica Callahan. Deep stories, nostalgia narratives, and fake news: Storytelling in the trump era. American Journal of Cultural Sociology, 5(3):392–408, 2017.
[141] Stephen Prochaska, Kelsey Duskin, Zhenya Kharazian, Caitlin Minow, Sam Blucker, Sarah Venuto, Jevin D. West, and Kate Starbird. Mobilizing manufactured reality: How participatory disinformation shaped deep stories to catalyze action during the 2020 u.s. presidential election. Proceedings of the ACM on Human-Computer Interaction, 7(CSCW1), 2023.
[142] V. Propp. Morphology of the Folktale. University of Texas Press, 1968.
[143] T. Radicioni, T. Squartini, E. Pavan, and F. Saracco. Networked partisanship and framing: A socio-semantic network analysis of the italian debate on migration. PLOS ONE, 16(8):1–24, 2021.
[144] Pranav Rajpurkar, Robin Jia, and Percy Liang. Know what you don’t know: Unanswerable questions for squad. CoRR, abs/1806.03822, 2018.
[145] H. Rashkin, E. Choi, J. Y. Jang, S. Volkova, and Y. Choi. Truth of varying shades: Analyzing language in fake news and political fact-checking. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 2931–2937, 2017.
[146] M. Reddi, R. Kuo, and D. Kreiss. Identity propaganda: Racial narratives and disinformation. New Media & Society, 25(8):2201–2218, 2023.
[147] N. Reimers and I. Gurevych. Reporting score distributions makes a difference: Performance study of lstm-networks for sequence tagging. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 338–348, 2017.
[148] M. Repnikova and K. Fang. Authoritarian participatory persuasion 2.0: Netizens as thought work collaborators in china. Journal of Contemporary China, 27(113):763–779, 2018.
[149] L. Rheault and A. Musulan. Eﬀicient detection of online communities and social bot activity during electoral campaigns. Journal of Information Technology & Politics, 18(3):324–337, 2021.
[150] K. Andrew R. Richards and Michael A. Hemphill. A practical guide to collaborative qualitative data analysis. Journal of Teaching in Physical Education, 37(2):225–231, 2018.
[151] Thomas Rid and Ben Buchanan. Attributing Cyber Attacks. Journal of Strategic Studies, 38(1-2):4–37, 2015.
[152] Thomas Rid and Ben Buchanan. Attributing Cyber Attacks. Journal of Strategic Studies, 38(1-2):4–37, 2015.
[153] M. Rosvall and C. T. Bergstrom. Maps of random walks on complex networks reveal community structure. Proceedings of the National Academy of Sciences of the USA, 105(4):1118–1123, 2008.
[154] Ivan Sanchez, Jeff Mitchell, and Sebastian Riedel. Behavior analysis of NLI models: Uncovering the influence of three factors on robustness. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), pages 1975–1985, New Orleans, Louisiana, June 2018. Association for Computational Linguistics.
[155] B. F. Schaffner and S. Luks. Misinformation or expressive responding? what an inauguration crowd can tell us about the source of political misinformation in surveys. Public Opinion Quarterly, 82(1):135–147, 2018.
[156] Max Schliebs, Hannah Bailey, Jonathan Bright, and Philip N. Howard. China＇ s public diplomacy operations: Understanding engagement and inauthentic amplification of prc diplomats on facebook and twitter. Oxford Internet Institute, 2021.
[157] Ozgur Can Seckin, Manita Pote, Alexander C Nwala, Lake Yin, Luca Luceri, Alessandro Flammini, and Filippo Menczer. Labeled Datasets for Research on Information Operations. Proceedings of the International AAAI Conference on Web and Social Media, 19(1):2567–2574, Jun. 2025.
[158] 中國新聞社 (China News Service). 【調查報告：新疆少數民族民眾勞動就業呈現明顯的自願性、自主性和自由性】新疆發展研究中心 20 日發布《新疆少數民族勞動就業調查報告》，顯示新疆少數民族民眾勞動就業呈現明顯的自願性、自主性和自由性。([survey report: The employment of ethnic minorities in xinjiang shows obvious voluntariness, autonomy and freedom] the xinjiang development research center released the ”survey report on the employment of ethnic minorities in xinjiang” on the 20th, showing that the employment of ethnic minorities in xinjiang shows obvious voluntariness, autonomy and freedom.), October 2020.
[159] 中國新聞社 (China News Service). 盛夏時節，新疆庫爾勒市連續多天出現晴好天氣，在夜晚，浩瀚星空下顯得格外神秘，樹林、高樓、燈光、星空構成一幅幅美麗的畫卷。# 魅力中國 (in midsummer, korla city in xinjiang has had sunny weather for several consecutive days. at night, the vast starry sky looks particularly mysterious, with trees, tall buildings, lights and the starry sky forming beautiful pictures. #charming china), August 2020.
[160] 中國新聞社 (China News Service). “上海人叫我｀小新疆＇，現在變成｀老新疆＇了。＂一口流利的上海話，不看長相以為就是地道的上海人……來自新疆喀什的艾尼到上海打拼已有 35 個年頭。憑著勤奮和聰明，如今的艾尼在上海虹口區買下了一棟四層沿街小樓做新疆餐廳，員工都來自新疆老家。來聽聽他的生意經？# 中國故事 (”shanghainese people used to call me ’little xinjiang’, but now they call me ’old xinjiang’.” he speaks fluent shanghainese and you would think he is a native shanghainese if you don’t look at his appearance... aini, who is from kashgar, xinjiang, has been working in shanghai for 35 years. with his diligence and intelligence, aini now bought a four-story street-side building in hongkou district, shanghai to run a xinjiang restaurant. all his employees are from xinjiang. want to hear his business tips? #chinastory), May 2021.
[161] 中國新聞社 (China News Service). “忽如一夜春風來，千樹萬樹梨花開＂。一夜之間，新疆北部布爾津縣被霧凇裝扮成童話仙境，行人駐足流連，親人約伴，共享、共賞新疆詩情畫意的冬季。# 魅力中國 (”suddenly, a spring breeze comes, and thousands of pear trees blossom.” overnight, burqin county in northern xinjiang was transformed into a fairytale wonderland by hoarfrost. pedestrians stopped to linger, and relatives made appointments to share and appreciate xinjiang’s poetic and picturesque winter. #charming china), December 2021.
[162] 中國新聞社 (China News Service). 【人均喜劇人盤點奧運健兒訪港的歡樂瞬間】受香港特區政府邀請，12 月 3 日到 5 日，內地奧運健兒訪港代表團到訪香港，展開了為期三天的交流活動，现场也是欢乐不断...... ([comedians per capita: A review of the joyful moments of the olympic athletes＇visit to hong kong] invited by the hong kong sar government, the mainland olympic athletes＇ delegation visited hong kong from december 3 to 5 and launched a three-day exchange activity. the scene was also full of joy...), December 2021.
[163] 中國新聞社 (China News Service). 【新疆博湖縣辣椒豐收農民樂】八月，新疆博湖縣迎來辣椒豐收季，村民們搶抓晴好天氣採摘、晾曬辣椒，田間地頭的曬場上一派繁忙景象。近年來，新疆博湖縣不斷調整優化農業產業結構，通過“合作社 + 基地 + 農戶＂模式進行訂單種植，有效帶動當地村民增收致富，助推鄉村振興。# 明天會更好 ([farmers are happy about the pepper harvest in bohu county, xinjiang] in august, the pepper harvest season came to bohu county, xinjiang. villagers took advantage of the fine weather to pick and dry the peppers, and the drying yards in the fields were bustling with activity. in recent years, bohu county, xinjiang has continuously adjusted and optimized its agricultural industrial structure, and carried out order planting through the ”cooperative + base + farmers” model, effectively driving local villagers to increase their income and become rich, and promoting rural revitalization. #tomorrow will be better), August 2021.
[164] 中國新聞社 (China News Service). 【新疆沙漠之門景區入選國家體育旅遊示範基地】近日，新疆塔克拉瑪干• 沙漠之門景區入選首批中國國家體育旅遊示範基地名單。沙漠之門景區於 2013 年開發建設，2021 年 5 月正式成為中國 5a 級景區，因其地處中國最大沙漠、世界第二大流動沙漠——塔克拉瑪干沙漠邊緣，故有沙漠之門之稱，也因此廣受關注。([xinjiang desert gate scenic area selected as a national sports tourism demonstration base] recently, xinjiang taklimakan desert gate scenic area was selected into the first batch of china’s national sports tourism demonstration bases. the desert gate scenic area was developed and constructed in 2013 and oﬀicially became a 5a-level scenic area in china in may 2021. because it is located on the edge of the taklimakan desert, the largest desert in china and the second largest mobile desert in the world, it is known as the desert gate and has therefore attracted widespread attention.), December 2021.
[165] 中國新聞社 (China News Service). 【氣勢磅礴！航拍新疆第一高橋自駕新疆不能錯過】新疆果子溝大橋建於群山之中，全長 700 米，橋面距谷底淨高達 200 餘米，是新疆第一座斜拉高橋。大橋蜿蜒曲折穿行在崇山峻岭之間，與周圍的美景渾然一體，成為眾多遊客自駕新疆的必到打卡地之一。# 魅力中國 ([magnificent!] aerial photography of the tallest bridge in xinjiang. don＇t miss it when driving in xinjiang. xinjiang guozigou bridge is built among the mountains. it is 700 meters long and the bridge deck is more than 200 meters above the valley bottom. it is the first cable-stayed bridge in xinjiang. the bridge winds its way through the mountains and blends seamlessly with the surrounding beautiful scenery, making it one of the must-visit spots for many tourists driving to xinjiang. #charming china), May 2021.
[166] 中國新聞社 (China News Service). 【汪文斌駁斥涉香港國安法錯誤言論】針對 “大赦國際＂組織涉香港國安法的錯誤言論，中國外交部發言人汪文斌 30 日駁斥道，“大赦國際＂組織的有關言論，純屬惡意抹黑歪曲之詞。任何對香港國安法的攻擊抹黑都阻擋不了香港由亂轉治、由治及興的大勢。([wang wenbin refutes erroneous remarks regarding the hong kong national security law] in response to the erroneous remarks made by amnesty international regarding the hong kong national security law, chinese foreign ministry spokesperson wang wenbin refuted on the 30th that the relevant remarks made by amnesty international were purely malicious smears and distortions. any attacks and smears on the hong kong national security law cannot stop the trend of hong kong’s transition from chaos to governance, and from governance to prosperity.), June 2021.
[167] 中國新聞社 (China News Service). 【涉非法集會反中亂港分子黎智英被判處 13 個月監禁】13 日，涉 2020 年一宗非法集會，香港 8 名人士煽惑他人參與未經批准集結等罪罪成，案件 13 日在香港區域法院判刑，8 人分別被判處 4 個月兩星期至 14 個月不等。經審訊後被裁定罪成的被告，壹傳媒創辦人黎智英被判處 13 個月監禁。([involved in illegal assembly, anti-china and hong kong disruptor jimmy lai was sentenced to 13 months in prison] on the 13th, eight people in hong kong were convicted of inciting others to participate in unauthorized assemblies in connection with an illegal assembly in 2020. the case was sentenced in the hong kong district court on the 13th. the eight people were sentenced to between 4 months and two weeks and 14 months. the defendant, jimmy lai, founder of next media, was found guilty after trial and sentenced to 13 months in prison.), December 2021.
[168] 中國新聞社 (China News Service). 【# 直播新疆特種旅遊節】新疆特種旅遊節在和田啟動。將播出宣傳片《出發吧！新疆》和田篇、新玉歌舞團鼓樂表演《龍騰虎躍》、歌舞表演《于闐樂舞》和《崑崙之夢》等精彩節目，更將發布和田冬春旅遊線路視頻。讓我們靜待春暖花開，一起相聚和田吧 ([#live broadcast xinjiang special tourism festival] xinjiang special tourism festival was launched in hotan. the promotional video ”let’s go! xinjiang” hotan chapter, the xinyu song and dance troupe’s drum and music performance ”dragons and tigers leaping”, the song and dance performances ”yarkand music and dance” and ”dream of kunlun” and other exciting programs, and the release of hotan winter and spring tourism route video. let us wait for the spring to come and gather together in hotan), December 2021.
[169] 中國新聞社 (China News Service). 【# 直播跨年夜逛和田夜市：新疆風味“亞克西＂】要說新疆和田夜市有什麼特色，不妨聽聽《和田夜市亞克西》。這裡除了 365 日飄香的鴿子湯、“三蛋一星＂、薄皮包子……人們更是歌不斷舞不停，到處散發着難以抵抗的“美食＂誘惑。跨年夜，在和田夜市的熱鬧氛圍中，品嘗地道新疆風味。([#live broadcast new year’s eve visit to hotan night market: Xinjiang flavor ”yaxi”] if you want to know what the characteristics of xinjiang hotan night market are, you might as well listen to ”hotan night market yaxi”. in addition to the pigeon soup, ”three eggs and one star”, and thin-skinned buns that are fragrant 365 days a year... people are singing and dancing non-stop, and the irresistible ”delicacies” temptation is everywhere. on new year＇ s eve, enjoy authentic xinjiang flavors in the lively atmosphere of hotan night market.), December 2021.
[170] 中國新聞社 (China News Service). 【香港國安法一周年特區各界全力落實】香港國安法即將迎來實施一周年。一年來，香港特區政府、香港特區維護國家安全委員會 (國安委)、各大政團等推出各項舉措落實香港國安法，並獲得社會各界響應。我們從時間線和社會各層面進行了梳理, 2021.
[171] 中國新聞社 (China News Service). 【香港特區第七屆立法會選舉 90 個議席全部產生】12 月 20 日晨，繼選舉選委會界別 40 個議席產生之後，香港特區第七屆立法會選舉分區直接選舉的 20 個議席及功能團體界別選舉的 30 個議席全部產生。至此，第七屆立法會 90 個議席全部順利產生。([all 90 seats in the seventh legislative council election of the hong kong special administrative region have been elected] on the morning of december 20, after the 40 seats in the election committee sector were elected, the 20 seats in the direct elections in the geographical constituencies and the 30 seats in the functional constituency elections in the seventh legislative council election of the hong kong special administrative region have all been elected. at this point, all 90 seats of the seventh legislative council have been successfully elected.), December 2021.
[172] 中國新聞社 (China News Service). 中國常駐聯合國代表團發言人 12 日就美、英、德夥同個別國家和非政府組織舉辦所謂中國新疆人權狀況邊會表示強烈譴責和堅決反對。發言人指出，這場邊會充斥着謊言和虛假信息，進一步暴露了美國等少數國家以疆制華、搞亂中國的政治圖謀，是一場徹頭徹尾的政治鬧劇 (on the 12th, the spokesperson of china’s permanent mission to the united nations expressed strong condemnation and resolute opposition to the socalled side event on human rights situation in xinjiang, china, held by the united states, britain and germany in collusion with some individual countries and nongovernmental organizations. the spokesperson pointed out that this side event was full of lies and false information, which further exposed the political plot of the united states and a few other countries to use xinjiang to contain china and disrupt china. it was an outright political farce.), May 2021.
[173] 中國新聞社 (China News Service). 只有如此，香港才能長治久安 (only in this way can hong kong achieve long-term stability.), June 2021.
[174] 中國新聞社 (China News Service). 新疆長絨棉世界頂級。做衣被暖和、透氣、舒適，長年供不應求。為滿足國內需求，中國每年需進口 200 萬噸左右棉花，近年來積極拓展進口渠道，與多個重要棉花生產國合作，確保國內棉花供應鏈穩定。中國還建立棉花儲備制度，確保棉農無虞。而隨着中國棉花機械化採摘推廣，2019 年新疆棉機收比例就已經達 42%。(xinjiang long-staple cotton is the best in the world. the clothes and quilts made from it are warm, breathable and comfortable, and are in short supply all year round. to meet domestic demand, china needs to import about 2 million tons of cotton every year. in recent years, it has actively expanded import channels and cooperated with several important cotton producing countries to ensure the stability of the domestic cotton supply chain. china has also established a cotton reserve system to ensure the safety of cotton farmers. with the promotion of mechanized cotton picking in china, the proportion of cotton harvested by machines in xinjiang reached 42March 2021.
[175] 中國新聞社 (China News Service). 護香港撐阿, 2021.
[176] 中國新聞社 (China News Service). 針對所謂“維吾爾特別法庭＂擬於近期再次舉行“聽證會＂，中國外交部發言人趙立堅 9 日表示，無論它找多少“演員＂、搞多少場所謂“聽證會＂，它都是一個非法“法庭＂，註定竹籃打水一場空。中國新疆的發展只會越來越好，國際社會主張客觀、公正看待新疆的聲音也只會越來越多。(regarding the so-called ”uyghur special tribunal”’s plan to hold another ”hearing” in the near future, chinese foreign ministry spokesperson zhao lijian stated on the 9th that no matter how many ”actors” it finds or how many so-called ”hearings” it holds, it is an illegal ”court” and is doomed to fail. the development of china’s xinjiang will only get better and better, and there will be more and more voices in the international community advocating an objective and fair view of xinjiang.), September 2021.
[177] 中國新聞社 (China News Service). 黑衣人，香港不歡迎你的存在 (men in black, hong kong does not welcome your presence), June 2021.
[178] 中國新聞社 (China News Service). “新疆是個好地方＂圖片展在日內瓦舉行】當地時間 9 月 26 日，“新疆是個好地方＂圖片展開幕式在日內瓦舉行。新疆維吾爾自治區人大常委會黨組書記、主任肖開提• 依明表示，“新疆是個好地方＂圖片展通過大量圖片，展示了新疆尊重和保障人權的顯著成就，把一個真實的新疆介紹給全世界。([xinjiang is a good place] photo exhibition held in geneva] on september 26, local time, the opening ceremony of the ”xinjiang is a good place” photo exhibition was held in geneva. xiao kaiti yimin, secretary of the party leadership group and director of the standing committee of the people’s congress of the xinjiang uygur autonomous region, said that the ”xinjiang is a good place” photo exhibition, through a large number of pictures, showcases xinjiang’s remarkable achievements in respecting and protecting human rights, and introduces a real xinjiang to the world.), 2022.
[179] 中國新聞社 (China News Service). 【新疆博湖：鵝喉羚野外“漫步＂悠閒覓食】12 月 11 日，位於新疆維吾爾自治區巴音郭楞蒙古自治州博湖縣博斯騰湖鄉南部的戈壁灘上，成群的鵝喉羚在悠閒覓食，它們或駐足凝望、或嬉戲奔跑，成為冬季一道亮麗的風景線。([xinjiang bohu: Goitered gazelles ”stroll” in the wild and forage for food leisurely] on december 11, in the gobi desert in the south of bosteng lake township, bohu county, bayinguoleng mongol autonomous prefecture, xinjiang uygur autonomous region, groups of goitered gazelles were leisurely foraging for food. they either stopped to stare or ran and played, creating a beautiful landscape in winter.), December 2022.
[180] 中國新聞社 (China News Service). 【新疆恢復線下招聘活動】新疆日前恢復線下招聘活動，各地州縣市開展就業援助“暖心活動＂延續至 12 月底。自 12 月 10 日開展就業援助“暖心活動＂以來，新疆認定援助對象 15133 人，實現就業 13530 人，就業率達 89% ([xinjiang resumes offline recruitment activities] xinjiang recently resumed offline recruitment activities, and various prefectures, counties and cities have carried out ”heartwarming activities” for employment assistance that will continue until the end of december. since the launch of the ”heart-warming activity” for employment assistance on december 10, xinjiang has identified 15,133 people for assistance, 13,530 of whom have found employment, with an employment rate of 89
[181] 中國新聞社 (China News Service).【# 東西問 | 梁玉春：為什麼說新疆人口發展是新疆發展進步的寫照？】新中國成立以來的新疆人口發展史，是新疆發生歷史性巨變的發展進步史。新疆師範大學民族與社會高等研究院院長梁玉春通過解讀新疆人口發展情況，就“新疆人口發展是社會進步的必然結果和生動寫照＂進行闡釋。([#east-west questions | liang yuchun: Why is xinjiang＇ s population development a reflection of xinjiang＇ s development and progress?] the history of xinjiang＇s population development since the founding of new china is a history of development and progress in which xinjiang has undergone historic changes. liang yuchun, dean of the xinjiang normal university institute of nationalities and society, explained the situation of xinjiang’s population development and how ”xinjiang’s population development is the inevitable result and vivid portrayal of social progress.”), January 2022.
[182] 中國新聞社 (China News Service). 【祖力亞提·司馬義：新疆不存在任何形式的“強迫勞動＂】針對所謂新疆“強迫勞動＂言論，新疆大學副校長、馬克思主義學院院長祖力亞提·司馬義 19 日指出，新疆各民族群眾的勞動完全是自願的、體面的，根本不存在任何形式的“強迫勞動＂。([zuliyati simayi: There is no ”forced labor” in any form in xinjiang] in response to the so-called ”forced labor” in xinjiang, zuliyati simayi, vice president of xinjiang university and dean of the school of marxism, pointed out on the 19th that the labor of people of all ethnic groups in xinjiang is completely voluntary and decent, and there is no ”forced labor” in any form at all.), May 2022.
[183] 中國新聞社 (China News Service). 就美國國務卿布林肯稱中國政府持續對新疆維族穆斯林和其他少數民族犯下種族滅絕和反人類罪行，中國外交部發言人汪文斌 2 日表示，中國新疆地區過去 60 多年來，人口總量增長了 4 倍，維吾爾族人口總量從 220 萬增長到約 1200 萬。說新疆“種族滅絕＂，是徹頭徹尾的“世紀謊言＂，毫無道德可言 (in response to u.s. secretary of state blinken’s statement that the chinese government continues to commit genocide and crimes against humanity against uyghur muslims and other ethnic minorities in xinjiang, chinese foreign ministry spokesperson wang wenbin said on the 2nd that the total population of china’s xinjiang region has increased fourfold over the past 60 years, and the total uyghur population has increased from 2.2 million to about 12 million. saying that xinjiang is ”genocide” is a complete ”lie of the century” and has no morals at all), March 2022.
[184] S. Shaar, N. Babulkov, G. Da San Martino, and P. Nakov. That is a known lie: Detecting previously fact-checked claims. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 3607–3618, 2020.
[185] Giuseppe Siracusano, Davide Sanvito, Roberto Gonzalez, Manikantan Srinivasan, Sivakaman Kamatchi, Wataru Takahashi, Masaru Kawakita, Takahiro Kakumaru, and Roberto Bifulco. Time for Action: Automated Analysis of Cyber Threat Intelligence in the Wild. CoRR, abs/2307.10214, 2023.
[186] C. B. Smith. The semantic attack surface: A systems-dynamic model of narrative in cyberspace. IEEE Transactions on Technology and Society, 4(2):146–157, June 2023.
[187] V. Solopova, O.-I. Popescu, C. Benzmüller, and T. Landgraf. Automated multilingual detection of pro-kremlin propaganda in newspapers and telegram posts. Datenbank Spektrum, 23:5–14, 2023.
[188] Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, and Ruslan Salakhutdinov. Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15:1929–1958, 2014.
[189] Kate Starbird, Ahmer Arif, and Tom Wilson. Disinformation as collaborative work: Surfacing the participatory nature of strategic information operations. Proceedings of the ACM on Human-Computer Interaction, 3(CSCW), 2019.
[190] Kate Starbird, Ahmer Arif, and Tom Wilson. Disinformation as Collaborative Work: Surfacing the Participatory Nature of Strategic Information Operations. Proc. ACM Hum.-Comput. Interact., 3(CSCW), Nov. 2019.
[191] Kate Starbird, Renee DiResta, and Matt DeButts. Influence and improvisation: Participatory disinformation during the 2020 u.s. election. Social Media + Society, 9(2), 2023.
[192] A. Steiner, A. Kolesnikov, X. Zhai, R. Wightman, J. Uszkoreit, and L. Beyer. How to train your vit? data, augmentation, and regularization in vision transformers. arXiv, 2020.
[193] Blake E. Strom, Andy Applebaum, Doug P. Miller, Katie C. Nickels, Adam G. Pennington, and Cody B. Thomas. MITRE ATT&CK: Design and Philosophy. Technical Report MP180486, MITRE, 2018. Technical Report.
[194] H. M. Sunde, J. Ilan, and S. Sandberg. A cultural criminology of “new” jihad: Insights from propaganda magazines. Crime, Media, Culture, 17(2):271–287, 2021.
[195] Vasantha P Suresh, Giulia Nogara, Fabrício Cardoso, Stefano Cresci, Stefano Giordano, and Leonardo Luceri. Tracking Fringe and Coordinated Activity on Twitter Leading Up To the US Capitol Attack. In Proceedings of the International AAAI Conference on Web and Social Media, 2024.
[196] SJ Terp and Pablo Breuer. DISARM: a Framework for Analysis of Disinformation Campaigns. In 2022 IEEE Conference on Cognitive and Computational Aspects of Situation Management (CogSIMA), pages 1–8, Jun. 2022.
[197] M. Thompson, J. Padget, and S. Battle. Governing narrative events with tropes as institutional norms. In Artificial Life and Intelligent Agents Symposium, 2016.
[198] 通說 (TongShuo). 【“黃絲＂藝人何韻詩被捕，“港獨＂“台獨＂都跳出來了！】“港獨＂藝人何韻詩今晨（29 日）被香港警方國安處拘捕。事情傳到台灣後，台灣 “綠委＂一番話似乎暴露了其早有離港赴台的心思，只不過如今沒成行便被抓了！有網友諷刺，有“綠委＂爆料“助攻＂，何韻詩“這下保釋都不可能了！＂, 2021.
[199] 通說 (TongShuo). 【今天，舉英國旗的香港” 王婆” 被判入獄！】據港媒報導，曾多次持英國國旗出現在“修例風波＂現場、人稱“王婆婆＂的王鳳瑤，2019 年 1 月在香港高等法院接受安檢期間，用肩膀撞一名女保安員的胸口，她早前否認普通襲擊罪，今天（14 日）在東區裁判法院受審，法官裁定其罪名成立，判即時監禁 1 個月。, 2021.
[200] 通說 (TongShuo). 【快訊】 3 月 1 日下午，香港西九龍裁判法院開庭審理 47 人被控串謀顛覆國家政權罪案件。庭外有示威者聚集並高叫口號，警方對其舉起紫色旗和藍色旗警告。([breaking news] on the afternoon of march 1, the west kowloon magistrates＇courts in hong kong opened a trial of 47 people charged with conspiracy to subvert state power. outside the court, protesters gathered and chanted slogans, and the police raised purple and blue flags to warn them.), March 2021.
[201] 通說 (TongShuo). 【網絡熱圖】上面美國的移民收容中心，下面是新疆的職業培訓中心，究竟哪個才是集中營？（資料圖：上帝之鷹）, 2021.
[202] 通說 (TongShuo). 【香港新冠疫苗流動接種站啟用】12 月 3 日，香港特區政府啟用新冠疫苗流動接種站，讓市民方便快捷地接種新冠疫苗。當日早晨有不少市民前往首站大圍新翠邨的新冠疫苗流動接種站接種新冠疫苗。（香港中通社記者許其皓攝）, 2021.
[203] 通說 (TongShuo). 【# 香港海洋公園下園區將免費開放】海洋公園公司董事局主席劉鳴煒 18 日表示，香港海洋公園未來山下園區，從港鐵出口至深水灣海邊，會成免費入場的零售餐飲消閒地區，並引入不同伙伴營運；山上園區則按景點收費，園區內的舊設施或需退休，將引入第三者夥伴經營，園方只保留動物及少量機動遊戲等設施。([#hong kong ocean park’s lower park will be open to the public free of charge] ocean park corporation chairman lau ming-wai said on the 18th that the future lower park of hong kong ocean park, from the mtr exit to the deep water bay beach, will become a free-admission retail, dining and leisure area, and will introduce different partners for operation; the upper park will charge according to the scenic spot, and the old facilities in the park may need to be retired and will be operated by a third party partner. the park will only retain animals and a small number of amusement rides and other facilities.), January 2021.
[204] 通說 (TongShuo). 【# 李家超拜訪香港漁農界選委 # 大批市民到場支持】香港特區第六屆行政長官候選人李家超 3 日前往香港仔，拜訪香港漁民團體聯會的相關界別選委。有市民得知李家超到來，陸續趕來等候在會所外，記者現場所見馬路兩旁站滿人，其中有坐著輪椅的長者。（記者李明珠）, 2022.
[205] Stephen E. Toulmin. The Uses of Argument. Cambridge University Press, 1958.
[206] Kevin Tseng and Man-Kwan Shan. LLM Agent for Disinformation Detection Based on DISARM Framework. In Workshop Proceedings of the 18th International AAAI Conference on Web and Social Media, 2024.
[207] N. Vanetik, M. Litvak, E. Reviakin, and M. Tiamanova. Propaganda detection in russian telegram posts in the scope of the russian invasion of ukraine. In Proceedings of the 14th International Conference on Recent Advances in Natural Language Processing, pages 1162–1170, 2023.
[208] P. Vijayaraghavan and S. Vosoughi. Tweetspin: Fine-grained propaganda detection in social media using multi-view representations. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 3433–3448, 2022.
[209] Douglas N. Walton. Informal logic: a handbook of critical argumentation. Cambridge University Press, 2005.
[210] L. Wang, X. Shen, G. de Melo, and G. Weikum. Cross-domain learning for classifying propaganda in online contents. In Proceedings of the 2020 Truth and Trust Online Conference, pages 21–31, 2020.
[211] Alicia Wanless and Michael Berk. The audience is the amplifier: Participatory propaganda. In The SAGE Handbook of Propaganda. SAGE, 2019.
[212] Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. Chain-of-thought prompting elicits reasoning in large language models. In Advances in Neural Information Processing Systems, 2022.
[213] Adina Williams, Nikita Nangia, and Samuel R. Bowman. A broad-coverage challenge corpus for sentence understanding through inference. CoRR, abs/1704.05426, 2017.
[214] T. Wilson, K. Zhou, and K. Starbird. Assembling strategic narratives: Information operations as collaborative work within an online community. Proceedings of the ACM on Human-Computer Interaction, 2(CSCW), November 2018.
[215] Samuel C. Woolley and Philip N. Howard. Computational Propaganda: Political Parties, Politicians, and Political Manipulation on Social Media. Oxford University Press, 2018.
[216] B. Wu, C. Xu, X. Dai, A. Wan, P. Zhang, Z. Yan, M. Tomizuka, J. Gonzalez, K. Keutzer, and P. Vajda. Visual transformers: Token-based image representation and processing for computer vision. arXiv, 2020.
[217] Bichen Wu, Chenfeng Xu, Xiaoliang Dai, Alvin Wan, Peizhao Zhang, Zhicheng Yan, Masayoshi Tomizuka, Joseph Gonzalez, Kurt Keutzer, and Peter Vajda. Visual transformers: Token-based image representation and processing for computer vision, 2020.
[218] Xinhua. Xinjiang’s progress in human rights smashes lies, rumors, July 2021.
[219] Xinhua. Hong kong’s successes propelled by motherland’s unwavering support, June 2022.
[220] K.-C. Yang, O. Varol, H. Pik-Mai, and F. Menczer. Scalable and generalizable social bot detection through data selection. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 1096–1103. AAAI, 2020.
[221] Savvas Zannettou, Tristan Caulfield, Emiliano Di Cristofaro, Michalis Sirivianos, Gianluca Stringhini, and Jeremy Blackburn. Disinformation Warfare: Understanding State-Sponsored Trolls on Twitter and Their Influence on the Web. arXiv preprint arXiv:1801.09288, 2018.
[222] Albert Zhang and Danielle Cave. Smart Asian women are the new targets of CCP global online repression, Jun. 2022.
[223] C. Zhang. Propaganda through participation: Counterterrorism narratives in china. In Research Handbook on Political Propaganda, pages 257–269. Edward Elgar Publishing, UK, 2021.
[224] Meiqian Zhao, Chunhua Liu, Lu Liu, Yan Zhao, and Dong Yu. Blcu_nlp at semeval-2018 task 12: An ensemble model for argument reasoning based on hierarchical attention. In Proceedings of The 12th International Workshop on Semantic Evaluation, pages 1104–1108. Association for Computational Linguistics, 2018.
[225] Zihao Zhao, Eric Wallace, Shi Feng, Dan Klein, and Sameer Singh. Calibrate Before Use: Improving Few-shot Performance of Language Models. In Marina Meila and Tong Zhang, editors, Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proceedings of Machine Learning Research, pages 12697–12706. PMLR, Jul. 2021.
[226] R. Zhong and P. Mozur. How china’s outrage machine kicked up a storm over h&m. New York Times, March 2021.
[227] Raymond Zhong and Paul Mozur. How China＇ s Outrage Machine Kicked Up a Storm Over H&M, Mar. 2021.
[228] New China 中文. 江布拉克，冬景如画！降雪让新疆奇台县南部的江布拉克景区披上银装，雪山、森林、灌木、峡谷交相辉映，仿佛一幅幅立体水墨画。江布拉克景区位于天山北麓、准噶尔盆地东南缘。(jiangbulake, the winter scenery is picturesque! the snowfall has covered the jiangbulake scenic area in the south of qitai county, xinjiang with silver. the snow-capped mountains, forests, shrubs and canyons complement each other, like a series of three-dimensional ink paintings. jiangbulake scenic area is located at the northern foot of tianshan mountain and the southeastern edge of junggar basin.), December 2020.
[229] New China 中文. 新疆 135 岁老人去世：经历三个世纪一家六代同堂据新疆疏勒县委宣传部消息，12 月 16 日，新疆喀什地区疏勒县居民阿丽米罕·色依提在喀什去世，享年 135 岁。阿丽米罕·色依提生于 1886 年 6 月 25 日 (清光绪十二年)，是新疆喀什地区疏勒县库木西力克乡人。阿丽米罕·色依提生前经历了 3 个世纪，一家六代同堂。, 2021.
[230] New China 中文. 新的一天：香港立法會選舉迎來投票日 (a new day: Voting day for hong kong legislative council election begins), December 2021.
[231] New China 中文. 这位 30 岁的维吾尔族姑娘，曾受宗教极端思想影响，在母亲劝说下，去了职业技能教育培训中心。结业后的她，成为喀什市香妃园的舞蹈演员。她说跳舞给了她灿烂，她感觉特别自由！点开视频，听她告诉两个美国专家，56 个民族，每一个民族都是中国的心头肉！(this 30-year-old uyghur girl was once influenced by religious extremism and went to a vocational skills education and training center at the persuasion of her mother. after graduation, she became a dancer at the xiangfei garden in kashgar. she said dancing gave her radiance and she felt especially free! click on the video and listen to her tell two american experts that every one of the 56 ethnic groups is dear to china＇ s heart!), December 2021.
[232] New China 中文. 香港學子與航天員共築「奮鬥夢想」(hong kong students and astronauts work together to realize their ”dream of striving”), January 2021.
[233] New China 中文. 香港將軍澳大橋主橋合龍 (hong kong tseung kwan o bridge main bridge completed), December 2021.
[234] New China 中文. 香港市民：候選人政綱影響香港未來 (hong kong citizens: Candidates＇political platforms affect hong kong＇s future), December 2021.
[235] New China 中文. 香港第七屆立法會選舉充分彰顯新選制的廣泛代表性、政治包容性、均衡參與性和公平競爭性 (the seventh legislative council election in hong kong fully demonstrated the broad representation, political inclusiveness, balanced participation and fair competition of the new electoral system), December 2021.
[236] New China 中文. 马踏雪原，景象壮观。冬日，在新疆伊犁哈萨克自治州昭苏县，骏马奔驰在雪原上，成为当地冬季特有的风景。(the sight of horses riding on the snowy plain is spectacular. in winter, in zhaosu county, ili kazakh autonomous prefecture, xinjiang, horses galloping on the snowy fields have become a unique winter scenery in the area.), December 2021.
[237] New China 中文. 香港多所學校元旦舉行升旗儀式迎接新年 (many schools in hong kong held flag-raising ceremonies on new year’s day to welcome the new year), January 2022.

簡易檢索 / 詳目顯示

相關論文