Papers in arXiv's cs.* by Sridhar Mahadevan in 2025-26 that contain apparently fabricated or incorrect citations. For each such citation: the citation as printed in the paper, and either the correct citation (for malformed references) or what the cited identifier actually resolves to (for wrong identifiers). The first two papers might be pure copy/paste errors, but proof-reading would have picked up two different adjacent papers with identical arXiv referenecs. --- ## "Universal Reinforcement Learning in Coalgebras" 1. Wrong arXiv ID (ID belongs to a different paper by the same author) - Citation: Sridhar Mahadevan. Universal imitation games. CoRR, abs/2110.15431, 2021. URL https://arxiv.org/abs/2110.15431. - That ID resolves to a different paper: is Sridhar Mahadevan, Universal Decision Models (2021). - "Universal Imitation Games" is actually . --- ## "Consciousness as a Functor" 1. Wrong arXiv ID (same error as 2508.15128) - Citation: Sridhar Mahadevan. Universal imitation games. CoRR, abs/2110.15431, 2021. URL https://arxiv.org/abs/2110.15431. - That ID resolves to a different paper: is Sridhar Mahadevan, Universal Decision Models (2021). - "Universal Imitation Games" is actually . --- ## "Large Causal Models from Large Language Models" 1. Fabricated (no such paper) - Citation: Xiaomei He et al. A survey of event causality identification: Taxonomy, resources, and challenges. ACM Computing Surveys, 2025. Preprint - No paper of this title by Xiaomei He exists; the only ACM Computing Surveys paper of this name is Qing Cheng, Zefan Zeng, Xingchen Hu, Yuehang Si, Zhong Liu, A Survey of Event Causality Identification: Taxonomy, Challenges, Assessment, and Prospects (2026) . 2. Wrong arXiv ID (resolves to an unrelated paper) - Citation: Zhiting Jin et al. Causal inference and natural language processing: A review. arXiv preprint arXiv:2012.14366, 2021. - Resolves to: is Jian Zhang, Cunjing Ge, Feifei Ma, Counting the Number of Solutions to Constraints (2020). 3. Fabricated (arXiv ID, author, title do not exist) - Citation: Vladislav Kosoy et al. Do large language models have causal knowledge? arXiv preprint arXiv:2305.00000, 2023. - Resolution: is not a working arXiv link, the title does not resolve to any paper on a web search, and an arXiv author search returns no "Vladislav Kosoy", neither does a Google Scholar search. A researcher named Eliza Kosoy works in the area. 4. Wrong arXiv ID (resolves to an unrelated paper) - Citation: Tuan Le et al. Multi-agent causal discovery using large language models. arXiv preprint arXiv:2404.12345, 2024. - Resolves to: is Shuai Li, Ming Gong, Yu-Hang Li, et al., High spin axion insulator (2024). - The real paper with this title: (Hao Duong Le, Xin Xia, Haijie Xu, Chen Zhang), the first author "Tuan Le" and the ID 2404.12345 are both wrong. 5. Wrong arXiv ID (resolves to an unrelated paper) - Citation: Xinyu Shen et al. Large language models for causal discovery and inference: A survey. arXiv preprint arXiv:2309.12345, 2023. - Resolves to: is Luís Arandas, Mick Grierson, Miguel Carvalhais, Antagonising explanation and revealing bias directly through sequencing and multimodal inference (2023). --- ## "Categorical Belief Propagation" 1. Wrong arXiv ID + wrong authors - Citation: B. Fong and D. I. Spivak. A categorical approach to probability theory. arXiv preprint arXiv:1406.6030, 2014 (versioned; commonly cited 2019). - That ID resolves to a different author's paper: is Kirk Sturtz, Categorical Probability Theory (2014). - Fong and Spivak have no paper with this title (checked across arXiv and the web). The title 'A categorical approach to probability theory' is a different work: Michèle Giry (Lecture Notes in Mathematics, 1982), . --- ## "CSQL: Mapping Documents into Causal Databases" 1. Wrong first author - Citation: Abhimanyu Dubey et al. The llama 3 herd of models. arXiv preprint arXiv:2407.21783, 2024. - The first author of is Aaron Grattafiori, the second author is Abhimanyu Dubey (and then a giant list, hence the et al) 2. Wrong title - Citation: Prashant Garg and Thiemo Fetzer. Testing causal claims in economics. arXiv preprint arXiv:2501.06873, 2025. URL https://arxiv.org/abs/2501.06873. - Correct title: is Causal Claims in Economics (Garg & Fetzer, 2025); authors and ID correct, the leading "Testing" is not part of the real title in either v1 or v2. 3. Fabricated (arXiv ID, author, title do not exist) - Citation: Vladislav Kosoy et al. Do large language models have causal knowledge? arXiv preprint arXiv:2305.00000, 2023. - Resolution: is not a working arXiv link, the title does not resolve to any paper on a web search, and an arXiv author search returns no "Vladislav Kosoy", neither does a Google Scholar search. A researcher named Eliza Kosoy works in the area. 4. Wrong arXiv ID (resolves to an unrelated paper) - Citation: Tuan Le et al. Multi-agent causal discovery using large language models. arXiv preprint arXiv:2404.12345, 2024. - Resolves to: is Shuai Li, Ming Gong, Yu-Hang Li, et al., High spin axion insulator (2024). - The real paper with this title: (Hao Duong Le, Xin Xia, Haijie Xu, Chen Zhang). 5. Wrong arXiv ID (resolves to an unrelated paper) - Citation: Xinyu Shen et al. Large language models for causal discovery and inference: A survey. arXiv preprint arXiv:2309.12345, 2023. - Resolves to: is Luís Arandas, Mick Grierson, Miguel Carvalhais, Antagonising explanation and revealing bias directly through sequencing and multimodal inference (2023). 6. Fabricated (no such paper) - Citation: Xiaomei He et al. A survey of event causality identification: Taxonomy, resources, and challenges. ACM Computing Surveys, 2025. Preprint - No paper of this title by Xiaomei He exists; the only ACM Computing Surveys paper of this name is Qing Cheng, Zefan Zeng, Xingchen Hu, Yuehang Si, Zhong Liu, A Survey of Event Causality Identification: Taxonomy, Challenges, Assessment, and Prospects (2026) . --- ## "PROMETHEUS: Automating Deep Causal Research" 1. Fabricated (nonexistent DOI, wrong authors/year/volume) - Citation: Xiaomei He, Yi Guan, and Min Chen. A survey of event causality identification: Taxonomy, resources, and techniques. ACM Computing Surveys, 55(14s):1–35, 2023. doi:10.1145/3582128. - The cited DOI does not resolve: returns HTTP 404. - A real paper with a similar title: Qing Cheng, Zefan Zeng, Xingchen Hu, Yuehang Si, Zhong Liu, A Survey of Event Causality Identification: Taxonomy, Challenges, Assessment, and Prospects, ACM Comput. Surv. 58(3), Art. 59, 2026 (different authors, year, volume, and subtitle). 2. Wrong arXiv ID (resolves to an unrelated paper) - Citation: Zhijing Jin, Bernhard Schölkopf, Peter Spirtes, and Kun Zhang. Causal inference and natural language processing: A survey. arXiv preprint arXiv:2012.14366, 2021. - Resolves to: is Jian Zhang, Cunjing Ge, Feifei Ma, Counting the Number of Solutions to Constraints (2020). 3. Wrong author list (real paper, correct ID) - Citation: Hao Duong Le, Xin Xia, and Zhang Chen. Multi-agent causal discovery using large language models. arXiv preprint arXiv:2407.15073, 2024. - Correct author list: is by Hao Duong Le, Xin Xia, Haijie Xu, Chen Zhang, the citation omits Haijie Xu and renders "Chen Zhang" as "Zhang Chen". --- ## "Causal Density Functions" 1. No such paper (bare conference citation, no identifier) - Citation: Nan Ke, Alexander Didic, Xinyu Chen, Seungjin Kim, and Yoshua Bengio. Learning causal effects via causal inference theory. In International Conference on Learning Representations, 2019. - Resolution: no paper with this title exists - DBLP search returns nothing (), it is not in the ICLR 2019 OpenReview proceedings (), and it is not among Ke's ICLR 2019 papers. - Closest real paper: Nan Rosemary Ke, Olexa Bilaniuk, Anirudh Goyal, et al., Learning Neural Causal Models from Unknown Interventions - , 2019.