https://gptzero.me/news/neurips/ Toggle menu [solid-logo-2] * Search * Dashboard * All News * Education * Investigations * Technology * Pricing * About Us Search Search Dashboard Investigations Featured GPTZero finds 100 new hallucinations in NeurIPS 2025 accepted papers GPTZero's analysis 4841 papers accepted by NeurIPS 2025 show there are at least 100 with confirmed hallucinations [avat] Nazar Shmatko [avat] Alex Adam [avat] Paul Esau Nazar Shmatko, Alex Adam, Paul Esau Jan 21, 2026 * 25 min read Share on X Share on Facebook Share on Linkedin Fact checked Copy citation to this article Copy link Send by email [NeurIPS-logo] The Conference on Neural Information Processing Systems (NeurIPS) is one of the most prestigious AI conferences in the world. The most recent meeting occurred in November 2025, where 100+ hallucinated citations were published. Table of contents Last month, GPTZero used our Hallucination Check tool to uncover 50 hallucinated citations in papers under review for ICLR 2026. However, we knew that ICLR (the International Conference on Learning Representations) was just one of hundreds of academic conferences and publications besieged by a tsunami of AI slop. After scanning 4841 papers accepted by the equally prestigious Conference on Neural Information Processing Systems (NeurIPS), we discovered 100s of hallucinated citations missed by the 3+ reviewers who evaluated each paper. Below, we uncover 100 confirmed hallucinations in the table below, spanning over 51 NeurIPS papers, which were not previously reported. [Number-of-hallucinations-by-author-affiliation--1-]Figure 1: Distribution of hallucinations by author's institution. How we computed this: a paper with 2 hallucinations with any authors from University A and University B will count as 2 hallucinations and 1 paper with hallucinations for both universities, independent of the number of authors from either university. A Problem of Scale ICLR, NeurIPS, ICML, and AAAI are the top machine learning / artificial intelligence conferences in the world, drawing thousands of submissions and participants annually. However, a submission tsunami fueled by generative AI, paper mills, and publication pressure has strained these conferences' review pipelines to the breaking point. Between 2020 and 2025, submissions to NeurIPS increased more than 220% from 9,467 to 21,575. In response, organizers have had to recruit ever greater numbers of reviewers, resulting in issues of oversight, expertise alignment, negligence, and even fraud. Our purpose in publishing these results is to illuminate a critical vulnerability in the peer review pipeline, not criticize the specific organizers, area chairs, or reviewers who participated in NeurIPS 2025. Over the past several years NeurIPS has changed the review process several times to address problems created by submission volume and generative AI tools. Still, our results reveal the consequences of a system that leaves academic reviewers, editors, and conference organizers outnumbered and outgunned -- trying to protect the rigor of peer review against challenges it was never designed to defend against. Table of 100 Hallucinated Citations in Published Across 53 NeurIPS Papers These NeurIPS papers have already been accepted, presented live, and effectively published. Since NeurIPS 2025 had an acceptance rate for main track papers of 24.52%, each of these papers beat out 15,000 other papers despite containing one or more hallucinations. This is concerning, given that the NeurIPS LLM policy considers hallucinated citations to be grounds for a paper's rejection or revocation, similar to ICLR. We've scanned each paper for both hallucinated citations (Sources) and AI-generated text (AI). An "*" next to the scan indicates the paper is likely a mix of AI and human text, while "**" indicates the paper is likely AI-generated. +-------------------------------------------------------------------+ |Published Paper |GPTZero|Example of Verified |Comment | | |Scan |Hallucination | | |-------------------+-------+----------------------+----------------| | | |John Doe and Jane |Article with a | | | |Smith. Webvoyager: |matching title | |SimWorld: An |Sources|Building an end-to-end|exists here. | |Open-ended | |web agent with large |Authors are | |Simulator for | |multimodal models. |obviously | |Agents in Physical |AI |arXiv preprint |fabricated. | |and Social Worlds | |arXiv:2401.00001, |arXiv ID links | | | |2024. |to a different | | | | |article. | |-------------------+-------+----------------------+----------------| | | |John Smith and Jane | | | | |Doe. Deep learning | | | | |techniques for | | |Unmasking | |avatar-based | | |Puppeteers: | |interaction in virtual|No author or | |Leveraging |Sources|environments. IEEE |title match. | |Biometric Leakage | |Transactions on Neural|Doesn't exist in| |to Expose | |Networks and Learning |publication. URL| |Impersonation in |AI* |Systems, 32 |and DOI are | |AI-Based | |(12):5600-5612, 2021. |fake. | |Videoconferencing | |doi: 10.1109/ | | | | |TNNLS.2021.3071234. | | | | |URL https:// | | | | |ieeexplore.ieee.org/ | | | | |document/307123 | | |-------------------+-------+----------------------+----------------| | | |Min-Jun Lee and | | | | |Soo-Young Kim. | | | | |Generative adversarial| | | | |networks for | | |Unmasking | |hyper-realistic avatar| | |Puppeteers: | |creation. In |No author or | |Leveraging |Sources|Proceedings of the |title match. | |Biometric Leakage | |IEEE/CVF Conference on|Doesn't exist in| |to Expose | |Computer Vision and |publication. URL| |Impersonation in |AI* |Pattern Recognition |and DOI are | |AI-Based | |(CVPR), pages |fake. | |Videoconferencing | |1234-1243, 2022. doi: | | | | |10.1109/ | | | | |CVPR.2022.001234. URL | | | | |https:// | | | | |ieeexplore.ieee.org/ | | | | |document/00123 | | |-------------------+-------+----------------------+----------------| |SimWorld-Robotics: | |Firstname Lastname and|No title or | |Synthesizing | |Others. Drivlme: A |author match. | |Photorealistic and |Sources|large-scale |Potentially | |Dynamic Urban | |multi-agent driving |referring to | |Environments for | |benchmark, 2023. URL |this article, | |Multimodal Robot |AI* |or arXiv ID to be |but year is off | |Navigation and | |updated. |(2024) | |Collaboration | | | | |-------------------+-------+----------------------+----------------| |SimWorld-Robotics: | |Firstname Lastname and|No title or | |Synthesizing | |Others. Robotslang: |author match. | |Photorealistic and |Sources|Grounded natural |Potentially | |Dynamic Urban | |language for |referring to | |Environments for | |multi-robot object |this article, | |Multimodal Robot |AI* |search, 2024. To |but year is | |Navigation and | |appear. |totally off | |Collaboration | | |(2020). | |-------------------+-------+----------------------+----------------| |Efficient semantic | |Nuo Lou and et al. | | |uncertainty |Sources|Dsp: Diffusion-based |No title or | |quantification in | |span prediction for |author match and| |language models via| |masked text modeling. |arXiv ID is | |diversity-steered |AI |arXiv preprint |incomplete. | |sampling | |arXiv:2305.XXXX, 2023.| | |-------------------+-------+----------------------+----------------| |Efficient semantic | |A. Sahoo and et al. | | |uncertainty |Sources|inatk: Iterative noise|No title or | |quantification in | |aware text denoising. |author match and| |language models via| |arXiv preprint |arXiv ID is | |diversity-steered |AI |arXiv:2402.XXXX, 2024.|incomplete. | |sampling | | | | |-------------------+-------+----------------------+----------------| |Efficient semantic | |Sheng Shi and et al. | | |uncertainty |Sources|Maskgpt: Uniform |No title or | |quantification in | |denoising diffusion |author match and| |language models via| |for language. arXiv |arXiv ID is | |diversity-steered |AI |preprint |incomplete. | |sampling | |arXiv:2401.XXXX, 2024.| | |-------------------+-------+----------------------+----------------| | | |Asma Issa, George | | | | |Mohler, and John | | | | |Johnson. Paraphrase | | |Efficient semantic | |identification using | | |uncertainty |Sources|deep contextualized |No author or | |quantification in | |representations. In |title match. No | |language models via| |Proceedings of the |match in | |diversity-steered |AI |2018 Conference on |publication. | |sampling | |Empirical Methods in | | | | |Natural Language | | | | |Processing (EMNLP), | | | | |pp. 517-526, 2018. | | |-------------------+-------+----------------------+----------------| | | |Yi Tay, Kelvin Fu, Kai| | | | |Wu, Ivan Casanueva, | | | | |Jianfeng Liu, Byron | | | | |Wallace, Shuohang | | | | |Wang, Bajrang Singh, | | |Efficient semantic | |and Julian McAuley. |No exact author | |uncertainty |Sources|Reasoning with |or title match, | |quantification in | |heterogeneous graph |although this | |language models via| |representations for |title is close. | |diversity-steered |AI |knowledge-aware |No match in the | |sampling | |question answering. In|publication. | | | |Findings of the | | | | |Association for | | | | |Computational | | | | |Linguistics: ACL 2021,| | | | |pp. 3497-3506, 2021. | | |-------------------+-------+----------------------+----------------| | | |Alex Wang, Rishi | | | | |Bommasani, Dan | | | | |Hendrycks, Daniel | | |Efficient semantic | |Song, and Zhilin |No title or | |uncertainty |Sources|Zhang. Efficient |author match. | |quantification in | |fewshot learning with |ArXiv ID leads | |language models via| |efl: A single |to a different | |diversity-steered |AI |transformer for all |article. | |sampling | |tasks. In arXiv | | | | |preprint | | | | |arXiv:2107.13586, | | | | |2021. | | |-------------------+-------+----------------------+----------------| | | |Lei Yu, Jimmy Dumsmyr,| | | | |and Kevin Knight. Deep| | |Efficient semantic | |paraphrase | | |uncertainty |Sources|identification. In |No title or | |quantification in | |Proceedings of the |author match. No| |language models via| |2014 Conference on |match in | |diversity-steered |AI |Empirical Methods in |publication | |sampling | |Natural Language | | | | |Processing (EMNLP), | | | | |pp. $650-655,2014$. | | |-------------------+-------+----------------------+----------------| |Efficient semantic | |X. Ou and et al. | | |uncertainty |Sources|Tuqdm: Token unmasking| | |quantification in | |with quantized |No title or | |language models via| |diffusion models. In |author match. | |diversity-steered |AI |ACL, 2024. | | |sampling | | | | |-------------------+-------+----------------------+----------------| | | |Franz Aichberger, Lily| | |Efficient semantic | |Chen, and John Smith. | | |uncertainty |Sources|Semantically diverse |No title or | |quantification in | |language generation. |author match. | |language models via| |In International |Some similarity | |diversity-steered |AI |Conference on Learning|to this article | |sampling | |Representations | | | | |(ICLR), 2025. | | |-------------------+-------+----------------------+----------------| | | |Maria Glushkova, | | | | |Shiori Kobayashi, and | | | | |Junichi Suzuki. | | |Efficient semantic | |Uncertainty estimation| | |uncertainty |Sources|in neural text |No author or | |quantification in | |regression. In |title match. No | |language models via| |Findings of the |match in | |diversity-steered |AI |Association for |publication. | |sampling | |Computational | | | | |Linguistics: EMNLP | | | | |2021, pp. | | | | |$4567-4576,2021$. | | |-------------------+-------+----------------------+----------------| | | |Yichao Wang, Bowen | | | | |Zhou, Adam Lopez, and | | | | |Benjamin Snyder. | | |Efficient semantic | |Uncertainty | | |uncertainty |Sources|quantification in | | |quantification in | |abstractive |No author or | |language models via| |summarization. In |title match. | |diversity-steered |AI |Proceedings of the | | |sampling | |60th Annual Meeting of| | | | |the Association for | | | | |Computational | | | | |Linguistics (ACL), pp.| | | | |1234-1245, 2022. | | |-------------------+-------+----------------------+----------------| | | |Mohit Jain, Ethan | | | | |Perez, and James | | |Efficient semantic | |Glass. Learning to | | |uncertainty |Sources|predict confidence for|No author or | |quantification in | |language models. In |title match. No | |language models via| |Proceedings of the |match in | |diversity-steered |AI |2021 Conference on |publication | |sampling | |Empirical Methods in | | | | |Natural Language | | | | |Processing (EMNLP), | | | | |pp. 245-256, 2021. | | |-------------------+-------+----------------------+----------------| | | |Srinivasan Kadavath, | | |Efficient semantic | |Urvashi Khandelwal, | | |uncertainty |Sources|Alec Radford, and Noam|No author or | |quantification in | |Shazeer. Answer me |title match. | |language models via| |this: Self-verifying |ArXiv ID leads | |diversity-steered |AI |large language models.|to a different | |sampling | |In arXiv preprint |article. | | | |arXiv:2205.05407, | | | | |2022. | | |-------------------+-------+----------------------+----------------| | | |Zayne Sprague, Xi Ye, |Two authors are | | | |Kyle Richardson, and |omitted and one | |Privacy Reasoning |Sources|Greg Durrett. MuSR: |(Kyle | |in Ambiguous | |Testing the limits of |Richardson) is | |Contexts | |chain-of-thought with |added. This | | |AI |multistep soft |paper was | | | |reasoning. In EMNLP, |published at | | | |2023. |ICLR 2024. | |-------------------+-------+----------------------+----------------| | | |Mario Paolone, Trevor | | | | |Gaunt, Xavier | | | | |Guillaud, Marco |The authors | | | |Liserre, Sakis |match this paper| |Memory-Augmented | |Meliopoulos, Antonello|, but the title,| |Potential Field |Sources|Monti, Thierry Van |publisher, | |Theory: A Framework| |Cutsem, Vijay Vittal, |volume, issue, | |for Adaptive | |and Costas Vournas. A |and page numbers| |Control in |AI** |benchmark model for |are incorrect. | |Non-Convex Domains | |power system stability|Year (2020) is | | | |controls. IEEE |correct. | | | |Transactions on Power | | | | |Systems, 35 | | | | |(5):3627-3635, 2020. | | |-------------------+-------+----------------------+----------------| | | |Mingliang Han, Bingni | | | | |W Wei, Phelan Senatus,| | | | |Jorg D Winkel, Mason | | |Memory-Augmented | |Youngblood, I-Han Lee,|No title or | |Potential Field |Sources|and David J Mandell. |author match. | |Theory: A Framework| |Deep koopman operator:|Journal and | |for Adaptive | |A model-free approach |other | |Control in |AI** |to nonlinear dynamical|identifiers | |Non-Convex Domains | |systems. Chaos: An |match this | | | |Interdisciplinary |article. | | | |Journal of Nonlinear | | | | |Science, 30 | | | | |(12):123135, 2020. | | |-------------------+-------+----------------------+----------------| |Adaptive | |Francisco Ramalho, | | |Quantization in | |Meng Liu, Zihan Liu, | | |Generative Flow |Sources|and Etienne Mathieu. |No author or | |Networks for | |Towards gflownets for |title match. | |Probabilistic | |continuous control. |ArXiv ID matches| |Sequential |AI |arXiv preprint |this paper. | |Prediction | |arXiv:2310.18664, | | | | |2023. | | |-------------------+-------+----------------------+----------------| | | |Arjun Gupta, Xi | | | | |Victoria Lin, Chunyuan| | | | |Zhang, Michel Galley, | | | | |Jianfeng Gao, and | | | | |Carlos Guestrin |No title or | |Grounded |Sources|Ferrer. Robust |author match. | |Reinforcement | |compositional visual |This paper has a| |Learning for Visual| |reasoning via |similar title | |Reasoning |AI* |language-guided neural|and matches | | | |module networks. In |publication. | | | |Advances in Neural | | | | |Information Processing| | | | |Systems (NeurIPS), | | | | |2021. | | |-------------------+-------+----------------------+----------------| | | |Diederik P. Kingma and| | |MTRec: Learning to | |Jimmy Ba. Deepfm: a |Title matches | |Align with User |Sources|factorization-machine |this paper. | |Preferences via | |based neural network |Authors, date, | |Mental Reward | |for ctr prediction. In|and publisher | |Models |AI |International |match this | | | |Conference on Learning|paper. | | | |Representations, 2015.| | |-------------------+-------+----------------------+----------------| | | |Weijia Xu, Xing Niu, | | | | |and Marine Carpuat. | | | | |Controlling toxicity |Authors, | |Redefining Experts:|Sources|in neural machine |publisher and | |Interpretable | |translation. In |date match this | |Decomposition of | |Proceedings of the |paper. Title and| |Language Models for|AI* |2020 Conference on |page numbers | |Toxicity Mitigation| |Empirical Methods in |don't match. | | | |Natural Language | | | | |Processing (EMNLP), | | | | |pages 4245-4256, 2020.| | |-------------------+-------+----------------------+----------------| | | |Xiang Zhang, Xuehai | | | | |Wei, Xian Zhang, and | | | | |Xue Zhang. Adversarial| | |Redefining Experts:| |attacks and defenses | | |Interpretable |Sources|in toxicity detection:|No author or | |Decomposition of | |A survey. In |title match. | |Language Models for| |Proceedings of the |Doesn't exist in| |Toxicity Mitigation|AI* |2020 International |publication. | | | |Joint Conference on | | | | |Neural Networks | | | | |(IJCNN), pages 1-8. | | | | |IEEE, 2020. | | |-------------------+-------+----------------------+----------------| | | |Fenglin Ding, Debesh | | | | |Jha, Maria Hargestam, | | | | |Pal Halvorsen, Michael| | | | |A Riegler, Dag | | | | |Johansen, Ronny | | | | |Hansch, and Havard |No title or | |Self-supervised | |Stensland. Vits: |author match. | |Learning of |Sources|Vision transformer for|Proceedings from| |Echocardiographic | |video self-supervised |this conference | |Video | |pretraining of |are split into | |Representations via|AI* |surgical phase |volumes, but the| |Online Cluster | |recognition. In |citation doesn't| |Distillation | |International |have a volume | | | |Conference on Medical |number. | | | |Image Computing and | | | | |Computer-Assisted | | | | |Intervention, pages | | | | |293-302. Springer, | | | | |2022. | | |-------------------+-------+----------------------+----------------| | | |Humberto | | | | |Acevedo-Viloria, Juan | | | | |Martinez, and Maria | | | | |Garcia. Relational | | |PANTHER: Generative|Sources|graph convolutional |No author or | |Pretraining Beyond | |networks for financial|title match. | |Language for | |fraud detection. IEEE |Doesn't exist in| |Sequential User |AI* |Transactions on |the cited | |Behavior Modeling | |Knowledge and Data |publication. | | | |Engineering, 33 | | | | |(7):1357-1370, 2021. | | | | |doi: 10.1109/ | | | | |TKDE.2020.3007655. | | |-------------------+-------+----------------------+----------------| | | |Majid Zolghadr, Mohsen| | | | |Jamali, and Jiawei | | | | |Zhang. Diffurecsys: | | |PANTHER: Generative| |Diffusion-based | | |Pretraining Beyond |Sources|generative modeling |No author or | |Language for | |for sequential |title match. DOI| |Sequential User | |recommendation. |doesn't exist. | |Behavior Modeling |AI* |Proceedings of the ACM| | | | |Web Conference (WWW), | | | | |pages 2156-2165, 2024.| | | | |doi: 10.1145/ | | | | |3545678.3557899. | | |-------------------+-------+----------------------+----------------| | | |Bernd Kerbl, Thomas |Loosely matches | |LiteReality: |Sources|Muller, and Paolo |this article, | |Graphic-Ready 3D | |Favaro. Efficient 3d |but only one | |Scene | |gaussian splatting for|author and part | |Reconstruction from|AI |real-time neural |of the title | |RGB-D Scans | |rendering. In CVPR, |actually match. | | | |2022. 2, 3 | | |-------------------+-------+----------------------+----------------| | | |Punchana Khungurn, | | |LiteReality: |Sources|Edward H. Adelson, |No clear match. | |Graphic-Ready 3D | |Julie Dorsey, and |Two authors and | |Scene | |Holly Rushmeier. |the subject | |Reconstruction from|AI |Matching real-world |match this | |RGB-D Scans | |material appearance. |article. | | | |TPAMI, 2015. 6 | | |-------------------+-------+----------------------+----------------| | | |Ashish Kumar, Logan | | | | |Engstrom, Andrew | | | | |Ilyas, and Dimitris | | |When and How |Sources|Tsipras. Understanding|No title or | |Unlabeled Data | |self-training for |author match. | |Provably Improve | |gradient-boosted |Doesn't exist in| |In-Context Learning|AI |trees. In Advances in |publication. | | | |Neural Information | | | | |Processing Systems | | | | |(NeurIPS), volume 33, | | | | |pp. 1651-1662, 2020. | | |-------------------+-------+----------------------+----------------| | | |Chuang Fan, Shipeng | | | | |Liu, Seyed Motamed, | | | | |Shiyu Zhong, Silvio | | | | |Savarese, Juan Carlos | | |When and How |Sources|Niebles, Anima |This paper | |Unlabeled Data | |Anandkumar, Adrien |exists, but all | |Provably Improve | |Gaidon, and Stefan |the authors are | |In-Context Learning|AI |Scherer. Expectation |fabricated. | | | |maximization pseudo | | | | |labels. arXiv preprint| | | | |arXiv:2305.01747, | | | | |2023. | | |-------------------+-------+----------------------+----------------| | | |T. Qiao, W. Liu, Z. | | |DualCnst: Enhancing| |Xie, H. Xu, J. Lin, J.|No clear author | |Zero-Shot | |Huang, and Y. Yang, |or title | |Out-of-Distribution|Sources|"Clip-score: A robust |matches. Title | |Detection via | |scoring metric for |loosely matches | |Text-Image | |text-to-image |this article. | |Consistency in |AI* |generation," arXiv |ArXiv ID leads | |Vision-Language | |preprint |here. | |Models | |arXiv:2201.07519, | | | | |2022. | | |-------------------+-------+----------------------+----------------| | | |Yunwen Lei, Puyu Wang,| | | | |Yiming Ying, and | | |Optimal Rates for |Sources|Ding-Xuan Zhou. | | |Generalization of | |Optimization and |No title match. | |Gradient Descent | |generalization of |Authors match | |for Deep ReLU |AI |gradient descent for |this paper. | |Classification | |shallow relu networks | | | | |with minimal width. | | | | |preprint, 2024. | | |-------------------+-------+----------------------+----------------| | | |Uher, R., Goodman, R.,| | | | |Moutoussis, M., | | |GeoDynamics: A | |Brammer, M., Williams,| | |Geometric | |S.C.R., Dolan, R.J.: |No exact title | |State-Space Neural |Sources|Cognitive and neural |or author match.| |Network for | |predictors of response|Loose title | |Understanding Brain| |to cognitive |match with this | |Dynamics on |AI* |behavioral therapy for|article. Doesn't| |Riemannian | |depression: a review |exist in the | |Manifolds | |of the evidence. |journal volume | | | |Journal of Affective | | | | |Disorders 169, 94-104 | | | | |(2014) | | |-------------------+-------+----------------------+----------------| | | |Junyeong Lee, Yiseong | | | | |Kim, Seungju Park, and| | | | |Hyunjik Lee. | | | | |Softmatch: Addressing | | | |Scan |the quantity-quality |Title matches | |Robust Label | |trade-off in |this paper. No | |Proportions | |semi-supervised |match in NeurIPS| |Learning |AI* |learning. In Advances |volume 36. | | | |in Neural Information | | | | |Processing Systems | | | | |(NeurIPS), volume 36, | | | | |pages 18315-18327, | | | | |2023. | | |-------------------+-------+----------------------+----------------| | | |Z. Zhu, T. Yu, X. | | | | |Zhang, J. Li, Y. | | | | |Zhang, and Y. Fu. | | | | |Neuralrgb-d: Neural | | |NFL-BA: Near-Field |Sources|representations for |No author or | |Light Bundle | |depth estimation and |title match. | |Adjustment for SLAM| |scene mapping. In |Doesn't exist in| |in Dynamic Lighting|AI |Proceedings of the |publication. | | | |IEEE Conference on | | | | |Computer Vision and | | | | |Pattern Recognition, | | | | |2022. | | |-------------------+-------+----------------------+----------------| | | |Y. Zhang, M. Oswald, | | | | |and D. Cremers. | | |NFL-BA: Near-Field |Sources|Airslam: |No author or | |Light Bundle | |Illumination-invariant|title match. | |Adjustment for SLAM| |hybrid slam. In |Doesn't exist in| |in Dynamic Lighting|AI |International |publication. | | | |Conference on Computer| | | | |Vision (ICCV), pages | | | | |2345-2354, 2023. | | |-------------------+-------+----------------------+----------------| | | |Yihong Zhu, Junxian |No title or | | | |Li, Xianfeng Han, |author match. | | |Sources|Shirui Pan, Liang Yao,|This paper has a| |Geometric Imbalance| |and Chengqi Wang. |similar title, | |in Semi-Supervised | |Spectral contrastive |but there's no | |Node Classification|AI |graph clustering. In |match in the | | | |International |ICLR 2022 | | | |Conference on Learning|database. | | | |Representations, 2022.| | |-------------------+-------+----------------------+----------------| | | |Ming Zhong, Han Liu, | | | | |Weizhu Zhang, Houyu | | | |Sources|Wang, Xiang Li, |No author or | |Geometric Imbalance| |Maosong Sun, and Xu |title match. | |in Semi-Supervised | |Han. Hyperbolic and |Doesn't exist | |Node Classification|AI |spherical embeddings |publication. | | | |for long-tail | | | | |entities. In ACL, | | | | |pages 5491-5501, 2021.| | |-------------------+-------+----------------------+----------------| | | |Ye Gao, Robert Tardif,| | | | |Jiale Cao, and Tapio |Title and | |NUTS: Eddy-Robust | |Schneider. Artificial |publisher match | |Reconstruction of |Sources|intelligence |this article. | |Surface Ocean | |reconstructs missing |Issue, page | |Nutrients via | |climate information. |numbers, and | |Two-Scale Modeling |AI* |Nature Geoscience, |year match this | | | |17:158-164, 2024. doi:|article. DOI is | | | |10.1038/ |fabricated. | | | |s41561-023-01297-2. | | |-------------------+-------+----------------------+----------------| | | | |Authors have | | | | |frequently | | | |Etienne Pardoux and |published | | | |Alexander Yu |together on the | |NUTS: Eddy-Robust |Sources|Veretennikov. Poisson |"poisson | |Reconstruction of | |equation for |equation", but | |Surface Ocean | |multiscale diffusions.|this title | |Nutrients via |AI* |Journal of |doesn't match | |Two-Scale Modeling | |Mathematical Sciences,|any of their | | | |111(3):3713-3719, |publications. | | | |2002. |Doesn't exist in| | | | |publication | | | | |volume/issue. | |-------------------+-------+----------------------+----------------| | | |Charanpal D Mummadi, | | | | |Matthias Arens, and | | |Test-Time | |Thomas Brox. Test-time| | |Adaptation of | |adaptation for | | |Vision-Language |Sources|continual semantic |No title or | |Models for | |segmentation. In |author match. | |Open-Vocabulary | |Proceedings of the |Doesn't exist in| |Semantic |AI* |IEEE/CVF Conference on|publication. | |Segmentation | |Computer Vision and | | | | |Pattern Recognition | | | | |(CVPR), pages | | | | |11828-11837, 2021. | | |-------------------+-------+----------------------+----------------| | | |Jiacheng He, Zhilu | | | | |Zhang, Zhen Wang, and | | |Test-Time | |Yan Huang. Autoencoder| | |Adaptation of |Sources|based test-time |No author or | |Vision-Language | |adaptation for |title match. | |Models for | |semantic segmentation.|Doesn't exist in| |Open-Vocabulary |AI* |In Proceedings of the |publication. | |Semantic | |IEEE/CVF International| | |Segmentation | |Conference on Computer| | | | |Vision (ICCV), pages | | | | |998-1007, 2021. | | |-------------------+-------+----------------------+----------------| | | |M. Gong, F. Yu, J. | | | | |Zhang, and D. Tao. | | | | |Efficient $\ell_{p}$ | | |Global Minimizers |Sources|norm regularization | | |of p-Regularized | |for learning sparsity |No title or | |Objectives Yield | |in deep neural |author match. | |the Sparsest ReLU |AI |networks. IEEE | | |Neural Networks | |Transactions on Neural| | | | |Networks and Learning | | | | |Systems, 33(10): | | | | |$5381-5392,2022 | | |-------------------+-------+----------------------+----------------| | | | |The authors | | | |Mihail Stoian, Richard|match this | | | |Milbradt, and |article and the | | | |Christian Mendl. |title is | |SHAP Meets Tensor |Sources|NP-Hardness of Optimal|similar. | |Networks: Provably | |TensorNetwork |However, the | |Tractable | |Contraction and |year, publisher | |Explanations with |AI |Polynomial-Time |and other data | |Parallelism | |Algorithms for Tree |don't match. | | | |Tensor Networks. |This article | | | |Quantum, 6:e119, 2022.|didn't appear in| | | | |the 2022 Quantum| | | | |volume. | |-------------------+-------+----------------------+----------------| | | |Jianyu Xu, Wei Li, and| | |SHAP Meets Tensor |Sources|Ming Zhao. Complexity |No title or | |Networks: Provably | |of Optimal Tensor |author match. | |Tractable | |Network Contraction |Doesn't exist in| |Explanations with |AI |Sequences. Journal of |publication. | |Parallelism | |Computational Physics,| | | | |480:112237, 2023. | | |-------------------+-------+----------------------+----------------| | | |Patrick Esser, Robin | | | | |Rombach, and Bjorn |Authors match | |Learning World |Sources|Ommer. Structure-aware|this article. | |Models for | |video generation with |ArXiv ID leads | |Interactive Video | |latent diffusion |to a different | |Generation |AI |models. arXiv preprint|article. | | | |arXiv:2303.07332, | | | | |2023. | | |-------------------+-------+----------------------+----------------| |Fourier Token | |Lele Xu, Chen Lin, | | |Merging: | |Hongyu Zhao, and et | | |Understanding and |Sources|al. Gaborvit: Global |No author or | |Capitalizing | |attention with local |title match. No | |Frequency Domain | |frequency awareness. |match in | |for Efficient Image|AI |In European Conference|publication. | |Generation | |on Computer Vision | | | | |(ECCV), 2022. | | |-------------------+-------+----------------------+----------------| | | |Yoonwoo Lee, Jaehyeong| | | | |Kang, Namil Kim, | | |Fourier Token | |Jinwoo Shin, and | | |Merging: | |Honglak Lee. | | |Understanding and |Sources|Structured fast |No author or | |Capitalizing | |fourier transform |title match. | |Frequency Domain | |attention for vision |Doesn't exist in| |for Efficient Image|AI |transformers. In |publication. | |Generation | |Advances in Neural | | | | |Information Processing| | | | |Systems (NeurIPS), | | | | |2022. | | |-------------------+-------+----------------------+----------------| | | |Siyuan Gong, Alan Yu, | | | | |Xiaohan Chen, Yinpeng | | |Fourier Token | |Lin, and Larry S | | |Merging: |Sources|Davis. Vision |No author or | |Understanding and | |transformer |title match. No | |Capitalizing | |compression: Early |match in | |Frequency Domain |AI |exiting and token |publication. | |for Efficient Image| |pruning. In Advances | | |Generation | |in Neural Information | | | | |Processing Systems | | | | |(NeurIPS), 2021. | | |-------------------+-------+----------------------+----------------| | | |Jiuxiang Shi, Zuxuan | | |Fourier Token | |Wu, and Dahua Lin. | | |Merging: |Sources|Token-aware adaptive |No author or | |Understanding and | |sampling for efficient|title match. | |Capitalizing | |diffusion models. In |Doesn't exist in| |Frequency Domain |AI |IEEE/CVF Conference on|publication. | |for Efficient Image| |Computer Vision and | | |Generation | |Pattern Recognition | | | | |(CVPR), 2023. | | |-------------------+-------+----------------------+----------------| | | |Raphael Muller, Simon | | |Fourier Token | |Kornblith, and | | |Merging: | |Geoffrey Hinton. |Authors match | |Understanding and |Sources|Adavit: Adaptive |this article. | |Capitalizing | |tokens for efficient |Title matches | |Frequency Domain | |vision transformer. In|this article. No| |for Efficient Image|AI |Proceedings of the |match in | |Generation | |International |publication. | | | |Conference on Machine | | | | |Learning (ICML), 2021.| | |-------------------+-------+----------------------+----------------| | | |Xin Wang, Anlin Chen, | | | | |Lihui Xie, Xin Jin, | | |Fourier Token | |Cheng Wang, and Ping |No author or | |Merging: |Sources|Luo. Not all tokens |title match. | |Understanding and | |are equal: Efficient |This article | |Capitalizing | |transformer for |title is | |Frequency Domain |AI |tokenization and |similar. No | |for Efficient Image| |beyond. In Advances in|match in | |Generation | |Neural Information |publication. | | | |Processing Systems | | | | |(NeurIPS), 2021. | | |-------------------+-------+----------------------+----------------| | | |Z. Chen and N. | | |A Unified Stability| |Flammarion. When and | | |Analysis of SAM vs |Sources|why sam generalizes |No author or | |SGD: Role of Data | |better: An |title match. | |Coherence and | |optimization |ArXiv ID leads | |Emergence of |AI |perspective. arXiv |to a different | |Simplicity Bias | |preprint |paper. | | | |arXiv:2206.09267, | | | | |2022. | | |-------------------+-------+----------------------+----------------| | | |K. A. Sankararaman, S.| | | | |Sankararaman, H. | | | | |Pandey, S. Ganguli, | | | | |and F. Bromberg. The | | |A Unified Stability| |impact of neural |This paper is a | |Analysis of SAM vs |Sources|network |match, but all | |SGD: Role of Data | |overparameterization |authors but the | |Coherence and | |on gradient confusion |first (K. A. | |Emergence of |AI |and stochastic |Sankararaman) | |Simplicity Bias | |gradient descent. In |are fabricated. | | | |37th International | | | | |Conference on Machine | | | | |Learning (ICML), pages| | | | |8469-8479, 2020. | | |-------------------+-------+----------------------+----------------| | | | |No author given | |MaterialRefGS: | |Why Physically-Based |and title | |Reflective Gaussian|Sources|Rendering. |appears to be | |Splatting with | |Physically-based |garbled. | |Multi-view | |rendering. Procedia |Publisher, | |Consistent Material|AI |IUTAM, 13(127137):3, |issue, year, and| |Inference | |2015 . |pages match this| | | | |article. | |-------------------+-------+----------------------+----------------| | | |Pierre Casgrain, | | |Robust | |Anirudh Kulkarni, and | | |Reinforcement | |Nicholas Watters. |No title or | |Learning in |Sources|Learning to trade with|author match. | |Finance: Modeling | |continuous action |ArXiv ID matches| |Market Impact with | |spaces: Application to|a different | |Elliptic |AI |market making. arXiv |article. | |Uncertainty Sets | |preprint | | | | |arXiv:2303.08603, | | | | |2023. | | |-------------------+-------+----------------------+----------------| |Robust | |Z Ning and Y K Kwok. | | |Reinforcement |Sources|Q-learning for option |No author or | |Learning in | |pricing and hedging |title match. No | |Finance: Modeling | |with transaction |match in journal| |Market Impact with |AI |costs. Applied |volume/issue. | |Elliptic | |Economics, 52 | | |Uncertainty Sets | |(55):6033-6048, 2020. | | |-------------------+-------+----------------------+----------------| |Robust | |W L Chan and R O | | |Reinforcement |Sources|Shelton. Can machine |No author or | |Learning in | |learning improve delta|title match. No | |Finance: Modeling | |hedging? Journal of |match in journal| |Market Impact with |AI |Derivatives, $9(1): |volume/issue. | |Elliptic | |39-56,2001$. | | |Uncertainty Sets | | | | |-------------------+-------+----------------------+----------------| |Robust | |Petter N Kolm, | | |Reinforcement | |Sebastian Krugel, and |No author or | |Learning in |Sources|Sergiy V Zadorozhnyi. |title match. | |Finance: Modeling | |Reinforcement learning|There is no | |Market Impact with | |for optimal hedging. |volume 14 of | |Elliptic |AI |The Journal of |this journal. | |Uncertainty Sets | |Trading, 14(4):4-17, | | | | |2019. | | |-------------------+-------+----------------------+----------------| | | |Kyung Hyun Park, | | |Robust | |Hyeong Jin Kim, and | | |Reinforcement |Sources|Woo Chang Kim. Deep |No author or | |Learning in | |reinforcement learning|title match. | |Finance: Modeling | |for limit order |Publisher ID | |Market Impact with |AI |book-based market |matches this | |Elliptic | |making. Expert Systems|article. | |Uncertainty Sets | |with Applications, | | | | |169:114338, 2021. | | |-------------------+-------+----------------------+----------------| | | |Moonseop Han and | | | | |Elizabeth Qian. Robust| | |FlowMixer: A | |prediction of |No author or | |Depth-Agnostic |Sources|dynamical systems with|title match. | |Neural Architecture| |structured neural |Publisher ID | |for Interpretable | |networks: Long-term |matches this | |Spatiotemporal |AI |behavior and chaos. |article. | |Forecasting | |Physica D: Nonlinear | | | | |Phenomena, 427:133006,| | | | |2021. | | |-------------------+-------+----------------------+----------------| | | | |No title match, | | | | |although | | | |Bart De Schutter and |Schutter and | |FlowMixer: A | |Serge P Hoogendoorn. |Hoogendorn have | |Depth-Agnostic |Sources|Modeling and control |written or | |Neural Architecture| |of freeway traffic |coauthored | |for Interpretable | |flow by state space |several related | |Spatiotemporal |AI |neural networks. |papers (example | |Forecasting | |Neural Computing and |and example). | | | |Applications, 17 |Journal volume/ | | | |(2):175-185, 2008. |issue matches an| | | | |unrelated | | | | |article. | |-------------------+-------+----------------------+----------------| | | |Jaideep Pathak, Brian | | | | |R Hunt, Georg M Goerg,| | |FlowMixer: A | |and Themistoklis P | | |Depth-Agnostic |Sources|Sapsis. Data-driven |No author or | |Neural Architecture| |prediction of chaotic |title match. No | |for Interpretable | |dynamics: Methods, |match in journal| |Spatiotemporal |AI |challenges, and |volume. | |Forecasting | |opportunities. Annual | | | | |Review of Condensed | | | | |Matter Physics, | | | | |14:379-401, 2023. | | |-------------------+-------+----------------------+----------------| | | |Alejandro Guemes, | | | | |Stefano Discetti, and | | |FlowMixer: A | |Andrea Ianiro. | | |Depth-Agnostic |Sources|Coarse-grained |No title or | |Neural Architecture| |physics-based |author match. | |for Interpretable | |prediction of |Doesn't exist in| |Spatiotemporal |AI |three-dimensional |journal volume/ | |Forecasting | |unsteady flows via |issue. | | | |neural networks. | | | | |Science Advances, 7 | | | | |(46):eabj0751, 2021. | | |-------------------+-------+----------------------+----------------| | | |Jeongseung Park, | | | | |Minseon Yang, Minz Won| | | | |Park, and Geonseok | | | | |Lee. Diffsound: | | |BNMusic: Blending |Sources|Differential sound |No title or | |Environmental | |manipulation with a |author match. | |Noises into | |few-shot supervision. |Doesn't exist in| |Personalized Music |AI* |In Proceedings of the |publication. | | | |IEEE/CVF Conference on| | | | |Computer Vision and | | | | |Pattern Recognition | | | | |(CVPR) Workshops, | | | | |pages 1767-1775, 2021.| | |-------------------+-------+----------------------+----------------| | | |Wenxuan Sun, Tri Dao, | | | | |Hongyu Zhuang, Zihang | | |Towards Multiscale | |Dai, Albert Gu, and |ArXiv ID leads | |Graph-based Protein|Sources|Christopher D Manning.|to this article | |Learning with | |Llamba: Efficient llms|with a similar | |Geometric Secondary| |with mamba-based |title and one | |Structural Motifs |AI* |distillation. arXiv |matching author.| | | |preprint | | | | |arXiv:2502.14458, | | | | |2024. | | |-------------------+-------+----------------------+----------------| | | |Tri Dao, Shizhe Ma, | | | | |Wenxuan Sun, Albert | | | | |Gu, Sam Smith, Aapo |Two authors (Tri| | | |Kyrola, Christopher D |Dao and Albert | |Towards Multiscale |Sources|Manning, and |Gu), the arXiv | |Graph-based Protein| |Christopher Re. An |ID, and the year| |Learning with | |empirical study of |match this paper| |Geometric Secondary|AI* |state space models for|. However, the | |Structural Motifs | |large language |title is only a | | | |modeling. arXiv |partial match. | | | |preprint | | | | |arXiv:2406.07887, | | | | |2024. | | |-------------------+-------+----------------------+----------------| | | |Junyan Zhu, Chenyang |No author or | |Fourier Clouds: | |Li, Chao He, and et |title match. | |Fast Bias |Sources|al. Freematch: A |This paper title| |Correction for | |simple framework for |is very close, | |Imbalanced | |long-tailed |but it was | |Semi-Supervised |AI |semi-supervised |published by | |Learning | |learning. In NeurIPS, |ICLR 2023 not | | | |2021. |NeurIPS 2021. | |-------------------+-------+----------------------+----------------| |NormFit: A | |Yijie Zang et al. |No author or | |Lightweight |Scan |Fedclip: A federated |title match, | |Solution for | |learning framework for|although this | |Few-Shot Federated | |vision-language |title is close. | |Learning with |AI |models. In NeurIPS, |No match in | |Non-IID Data | |2023. |publication. | |-------------------+-------+----------------------+----------------| | | |Jiahui Liu and et al. |No author or | |AI-Generated Video |Sources|Tall-swin: Thumbnail |title match. A | |Detection via | |layout transformer for|paper with a | |Perceptual | |generalised deepfake |similar title | |Straightening |AI |video detection. In |appears in | | | |ICCV, 2023. |publication. | |-------------------+-------+----------------------+----------------| | | |Nitish Srivastava and |No title match, | |Multi-Expert | |Ruslan R |but authors have| |Distributionally |Sources|Salakhutdinov. |published | |Robust Optimization| |Discriminative |together | |for | |features for fast |previously ( | |Out-of-Distribution|AI* |frame-based phoneme |example). No | |Generalization | |classification. Neural|match in | | | |networks, 47:17-23, |publication. | | | |2013. | | |-------------------+-------+----------------------+----------------| | | |Anh Tuan Nguyen, | | | | |Shengping Li, and Chao| | |MIP against Agent: | |Qin. Multimodal | | |Malicious Image |Sources|adversarial |No author or | |Patches Hijacking | |robustness: Attack and|title match. | |Multimodal OS | |defense. IEEE |Doesn't exist in| |Agents |AI |Transactions on |publication. | | | |Pattern Analysis and | | | | |Machine Intelligence | | | | |(TPAMI), 2021. | | |-------------------+-------+----------------------+----------------| | | |Jack Lau, Ankan Gayen,| | | | |Philipp Tschandl, | | | | |Gregory A Burns, | | | | |Jiahong Yuan, Tanveer | | | | |SyedaMahmood, and |No author match.| | | |Mehdi Moradi. A |Title matches | |ACT as Human: |Sources|dataset and |another | |Multimodal Large | |exploration of models |hallucinated | |Language Model Data| |for understanding |citation in this| |Annotation with |AI* |radiology images |paper. Doesn't | |Critical Thinking | |through dialogue. In |exist in | | | |Proceedings of the |publication. | | | |Conference on | | | | |Empirical Methods in | | | | |Natural Language | | | | |Processing (EMNLP), | | | | |pages 2575-2584, 2018.| | |-------------------+-------+----------------------+----------------| | | |Yikai Zhang et al. | | |OCTDiff: Bridged | |"Text-to-Image | | |Diffusion Model for|Sources|Diffusion Models with |No author or | |Portable OCT | |Customized Guidance". |title match. | |Super-Resolution | |In: Proceedings of the|Doesn't exist in| |and Enhancement |AI |IEEE/CVF International|publication. | | | |Conference on Computer| | | | |Vision. 2023. | | |-------------------+-------+----------------------+----------------| | | |Author Song and | | | | |AnotherAuthor Zhang. | | | | |"Consistency in |No author or | |OCTDiff: Bridged | |Diffusion Models: |title match. | |Diffusion Model for|Sources|Improving Noise |This paper has a| |Portable OCT | |Embeddings". In: IEEE |similar title. | |Super-Resolution | |Transactions on |ArXiv ID leads | |and Enhancement |AI |Pattern Analysis and |to unrelated | | | |Machine Intelligence |paper. | | | |(2023). URL: https:// | | | | |arxiv.org/abs/ | | | | |2304.08787. | | |-------------------+-------+----------------------+----------------| | | | |Author is a | | | | |famous economist| | | |Claudia Goldin. |, but the title | |Strategic Costs of |Sources|Occupational choices |doesn't match | |Perceived Bias in | |and the gender wage |any of her | |Fair Selection | |gap. American Economic|works. Journal | | |AI |Review, 104 |and locators | | | |(5):348-353, 2014. |match this | | | | |unrelated | | | | |article. | |-------------------+-------+----------------------+----------------| | | |Olah, C., Elhage, N., |Most authors | | | |Nanda, N., Schiefer, |match this paper| |Linear Transformers|Sources|N., Jones, A., |, but the title,| |Implicitly Discover| |Henighan, T., and |publisher, and | |Unified Numerical | |DasSarma, N. (2022). |year are | |Algorithms |AI |Transformer circuits. |different. | | | |Distill, 7(3). https:/|Doesn't exist in| | | |/distill.pub/2022/ |publication. | | | |circuits/. | | |-------------------+-------+----------------------+----------------| | | |Nanda, N. (2023). |No title match. | |Linear Transformers|Sources|Progress in |Author may be | |Implicitly Discover| |mechanistic |Neel Nanda, who | |Unified Numerical | |interpretability: |wrote several | |Algorithms |AI |Reverse-engineering |similar articles| | | |induction heads in |in 2023. | | | |GPT-2. | | |-------------------+-------+----------------------+----------------| | | |J. Zhang and X. Li. | | |A Tri-Modal | |Multi-agent systems | | |Multi-Agent |Sources|for distributed |No author or | |Responsive | |problem solving: A |title match. | |Framework for | |framework for task |Doesn't exist in| |Comprehensive 3D |AI* |decomposition and |publication. | |Object Annotation | |coordination. Procedia| | | | |Computer Science, | | | | |55:1131-1138, 2015. | | |-------------------+-------+----------------------+----------------| | | |Erfan Aghasian, Shai | | |A Tri-Modal | |Avidan, Piotr Dollar, | | |Multi-Agent |Sources|and Justin Johnson. |No author or | |Responsive | |Hierarchical protocols|title match. | |Framework for | |for multi-agent 3d |Doesn't exist in| |Comprehensive 3D |AI* |scene understanding. |publication. | |Object Annotation | |In CVPR, pages | | | | |7664-7673, 2021. | | |-------------------+-------+----------------------+----------------| | | |Itay Hubara, Matthieu | | | | |Courbariaux, Daniel | | | | |Soudry, Rami El-Yaniv,| | |Learning Grouped | |and Yoshua Bengio. |Authors mostly | |Lattice Vector |Sources|Quantizing deep |match this paper| |Quantizers for | |convolutional networks|. Title matches | |Low-Bit LLM | |for efficient |this paper. | |Compression |AI* |inference: A |ArXiv ID matches| | | |whitepaper. arXiv |a third paper. | | | |preprint | | | | |arXiv:1612.01462, | | | | |2017. | | |-------------------+-------+----------------------+----------------| | | |Zhiqiang Wang, Chao | | | | |Zhang, Bing Li, Zhen | | |Learning Grouped |Sources|Xu, and Zhiwei Li. A |No author match.| |Lattice Vector | |survey of model |Title matches | |Quantizers for | |compression and |this paper. | |Low-Bit LLM |AI* |acceleration for deep |Doesn't exist in| |Compression | |neural networks. ACM |publication. | | | |Computing Surveys, 54 | | | | |(7):1-34, 2021. | | |-------------------+-------+----------------------+----------------| | | |Andrew Black et al. | | |PointMapPolicy: |Sources|Zero-shot skill |No title match. | |Structured Point | |composition with |ArXiv ID leads | |Cloud Processing | |semantic feature |to unrelated | |for Multi-Modal |AI* |fusion. arXiv preprint|paper. | |Imitation Learning | |arXiv:2310.08573, | | | | |2023. | | |-------------------+-------+----------------------+----------------| | | |Yufei Wu, Kiran | | | | |Alwala, Vivek | | | | |Ganapathi, Sudeep | | | | |Sharma, Yilun Chang, | | |PointMapPolicy: |Sources|Yicheng Zhang, Yilun |No author or | |Structured Point | |Zhou, et al. Susie: |title match. | |Cloud Processing | |Scaling up |ArXiv ID leads | |for Multi-Modal |AI* |instruction-following |to unrelated | |Imitation Learning | |policies for robot |article. | | | |manipulation. arXiv | | | | |preprint | | | | |arXiv:2402.17552, | | | | |2024. | | |-------------------+-------+----------------------+----------------| | | |Zhipeng Zhang, Chang | | | | |Liu, Shihan Wu, and | | | | |Yan Zhao. EST: Event | | | | |spatio-temporal | | | | |transformer for object| | |FLAME: Fast |Sources|recognition with event|No author or | |Long-context | |cameras. In ICASSP |title match. | |Adaptive Memory for| |2023 - 2023 IEEE |Doesn't exist in| |Event-based Vision |AI |International |publication. | | | |Conference on | | | | |Acoustics, Speech and | | | | |Signal Processing | | | | |(ICASSP), pages 1-5. | | | | |IEEE, 2023. | | |-------------------+-------+----------------------+----------------| | | |Daniel Gehrig, Mathias| | | | |Gehrig, John Monaghan,| | | | |and Davide Scaramuzza.| | | | |Recurrent vision |No author match,| |FLAME: Fast |Sources|transformers for dense|but this paper | |Long-context | |prediction. In |has a similar | |Adaptive Memory for| |Proceedings of the |title. Doesn't | |Event-based Vision |AI |IEEE/CVF International|exist in | | | |Conference on Computer|publication. | | | |Vision (ICCV) | | | | |Workshops, pages | | | | |3139-3148, 2021. | | |-------------------+-------+----------------------+----------------| | | |Qiyang Du, Ozan Sener,|No author or | |Diversity Is All | |and Silvio Savarese. |title match. | |You Need for |Sources|Agree to disagree: |Sener and | |Contrastive | |Adaptive learning with|Savarese have | |Learning: Spectral | |gradient disagreement.|published | |Bounds on Gradient |AI** |In Advances in Neural |together | |Magnitudes | |Information Processing|previously. | | | |Systems (NeurIPS), |Doesn't exist in| | | |2021. |publication. | |-------------------+-------+----------------------+----------------| | | |Longxuan Jing, Yu | | | | |Tian, Yujun Pei, | | | | |Yibing Shen, and | | | | |Jiashi Feng. | | |Diversity Is All | |Understanding | | |You Need for |Sources|contrastive |No author match.| |Contrastive | |representation |Title matches | |Learning: Spectral | |learning through |this paper. | |Bounds on Gradient |AI** |alignment and |Doesn't exist in| |Magnitudes | |uniformity on the |publication. | | | |hypersphere. In | | | | |International | | | | |Conference on Learning| | | | |Representations | | | | |(ICLR), 2022. | | |-------------------+-------+----------------------+----------------| | | |Yair Leviathan, |Title, | |TokenSwap: A | |Clemens Rosenbaum, and|publisher, and | |Lightweight Method |Sources|Slav Petrov. Fast |date match this | |to Disrupt | |inference from |paper, but all | |Memorized Sequences| |transformers via |authors except | |in LLMs |AI* |speculative decoding. |one surname | | | |In ICML, 2023. |(Leviathan) are | | | | |different. | |-------------------+-------+----------------------+----------------| | | |Wenwen Chang, Tal | | |TokenSwap: A | |Schuster, and Yann | | |Lightweight Method |Sources|LeCun. Neural surgery |No author or | |to Disrupt | |for memorisation: |title match. | |Memorized Sequences| |Locating and removing |Doesn't exist in| |in LLMs |AI* |verbatim recall |publication. | | | |neurons. In NeurIPS, | | | | |2024. | | |-------------------+-------+----------------------+----------------| |Learning the Wrong | |M. Garcia and A. | | |Lessons: |Sources|Thompson. Applications|No author or | |Syntactic-Domain | |of llms in legal |title match. | |Spurious | |document analysis. |Publication | |Correlations in |AI |Journal of Legal |doesn't exist. | |Language Models | |Technology, 7 | | | | |(1):50-65, 2024. | | |-------------------+-------+----------------------+----------------| | | |J. Smith and A. Patel.| | |Learning the Wrong | |Leveraging large | | |Lessons: |Sources|language models for |No author or | |Syntactic-Domain | |financial forecasting.|title match. | |Spurious | |International Journal |Publication | |Correlations in |AI |of Financial |doesn't exist. | |Language Models | |Technology, 9 | | | | |(2):101-115, 2024. | | |-------------------+-------+----------------------+----------------| |JADE: Joint |Sources|David Jones et al. |No author or | |Alignment and Deep | |Gpsa: Gene expression |title match. | |Embedding for | |and histology-based |Doesn't exist in| |Multi-Slice Spatial|AI* |spatial alignment. |publication. | |Transcriptomics | |Nature Methods, 2023. | | |-------------------+-------+----------------------+----------------| | | |Zhihao Chen, Hantao | | | | |Zhang, Yuhan Zhang, | | | | |Zhanlin Hu, Quanquan | | | | |Gu, Qing Zhang, and | | |JADE: Joint |Sources|Shuo Suo. Slat: a |No author or | |Alignment and Deep | |transformer-based |title match. | |Embedding for | |method for |Doesn't exist in| |Multi-Slice Spatial|AI* |simultaneous alignment|publication. | |Transcriptomics | |and clustering of | | | | |spatial | | | | |transcriptomics data. | | | | |Nature Communications,| | | | |14(1):5548, 2023. | | |-------------------+-------+----------------------+----------------| | | |Francois Baccelli, | | | | |Gerard H. Tache, and | | |Improved Regret |Sources|Etienne Altman. Flow |No author or | |Bounds for Linear | |complexity and |title match. | |Bandits with | |heavytailed delays in |Doesn't exist in| |Heavy-Tailed |AI |packet networks. |publication. | |Rewards | |Performance | | | | |Evaluation, 49 | | | | |(1-4):427-449, 2002. | | |-------------------+-------+----------------------+----------------| | | |Saravanan | | | | |Jebarajakirthy, Paurav| | |Improved Regret | |Shukla, and Prashant | | |Bounds for Linear |Sources|Palvia. Heavy-tailed |No author or | |Bandits with | |distributions in |title match. | |Heavy-Tailed | |online ad response: A |Doesn't exist in| |Rewards |AI |marketing analytics |publication. | | | |perspective. Journal | | | | |of Business Research, | | | | |124:818-830, 2021. | | |-------------------+-------+----------------------+----------------| | | |Mehdi Azabou, Micah | | |AutoSciDACT: | |Weber, Wenlin Ma, et | | |Automated | |al. Mineclip: |No author or | |Scientific |Sources|Multimodal neural |title match. | |Discovery through | |exploration of clip |ArXiv ID leads | |Contrastive | |latents for automatic |to unrelated | |Embedding and |AI |video annotation. |article. | |Hypothesis Testing | |arXiv preprint | | | | |arXiv:2210.02870, | | | | |2022. | | +-------------------------------------------------------------------+ --------------------------------------------------------------------- Is there a specific report or published article you think we should check for hallucinations? Submit Here Defining Hallucinated Citations Given the high stakes for both authors and publishers, GPTZero's Hallucination Check is engineered to be accurate, transparent, and cautious. It uses our AI agent, trained in-house, to flag any citations in a document that can't be found online. These flagged citations are not automatically hallucinations -- many archival documents or unpublished works can't be matched to an online source -- but they indicate which sources require further human scrutiny. As always, we recommend that a human confirm that flagged citation is an AI-generated fake instead of the result of a more conventional error. We define a vibe citation as a citation that likely resulted from the use of generative AI. Vibe citing results in errors common to LLM generations, but rare in human-written text, such as: 1. Combining or paraphrasing the titles, author(s), and/or locators from one or more real sources 2. Fabricating the author(s), title, URL/DOI, and/or container (ex. publisher, journal, conference) of a source 3. Modifying the author(s) or title of a source by extrapolating a first name from an initial, dropping and/or adding authors, or paraphrasing the title. Our definition excludes obvious spelling mistakes, dead URLs, missing locators, and other errors that are plausibly human. The following table shows the difference between a real citation, a flawed citation, and a hallucinated citation according to our methodology. The differences are highlighted in red. Real Citation Flawed Citation Hallucinated Citation Yann LeCun, Yoshua Y. LeCun, Y. Bengio, and Samuel LeCun Jackson Bengio, and Geoffrey Geoff Hinton. Deep . Deep learning. Hinton. Deep learning. leaning. nature, 521 Science & Nature: nature, 521:436-444, (7553):436-444, 2015. 23-45, 2021. 2015. A. Yang, B. Zhang, B. A. Yang, (missing Hui, B. Gao, B. Yu, C. author), B. Hui, B. Gao, A. Yang, B. Yang, C. Li, D. Liu, J. Tu, J. B. Yu, C. Li/, D. Liu, Yang, et al. Qwen Zhou, J. Lin, et al. J. Tu, J. Zhou, J. Lin, 3.5-mathematical Qwen2.5-math technical et al. Qwen 2. 5-math report for iterative report: Toward technical report: Toward model mathematical expert mathematical expert self-improvement. model via model via arXiv:2909.12233, self-improvement. self-improvement. arXiv 2024. arXiv:2409.12122, 2024. preprint arXiv:2409.12122, 2024. Like GPTZero's AI Detector, Hallucination Check has an extremely low false negative rate, so we catch 99 out of 100 flawed citations. Because our tool will flag any citation that can't be verified online, the false positive rate is higher. Vibe Citing Over the past few months, we've experimented with several names for an LLM-generated citation with fabricated elements. "Hallucinated citations" is too long, "hallucitations" too easily mistaken for a spelling error, and "fake citations" too morally charged. Recently, GPTZero's Head of Machine Learning, Alex Adams, coined the term "vibe citing" to describe the LLM tendency to derive or amalgamate real sources into uncanny imitations. "Vibe citing," like "vibe writing" or "vibe coding" produces citations that look accurate at first glance, but crumble under closer inspection. [star-history-2026121]Figure 2: Open-source projects to write research papers with AI are booming in popularity and illustrate the growth in vibe-citing. The bumps in April and September 2025 correspond to the paper submission deadlines for NeurIPS and ICLR 2025. GPTZero's analysis of 4841 of the 5290 papers accepted by NeurIPS 2025 indicates noticeable traces of AI authorship and hundreds of vibe citations. As always, each of the hallucinations presented here has been verified by a human expert. Surf the Tsunami with Hallucination Check Hallucination Check is the only tool of its kind, and provides an essential service at multiple points in the peer review pipeline. First, it allows authors to check their manuscripts for citation errors -- including common issues that can occur without LLM involvement like dead links or partial titles. Second, it greatly reduces the time and labor necessary for reviewers to check a submission's sources and identify possible vibe citing. Third, using Hallucination Check in combination with GPTZero's AI Detector allows editors and conference chairs to check for AI-generated text and suspicious citations at the same time, leading to faster and more accurate editorial decisions. After releasing our ICLR paper investigation we are now coordinating with the ICLR team to review future paper submissions. As always, our goal is to make the peer review process faster, fairer, and more transparent for everyone involved. Try GPTZero's Hallucination check for yourself, or reach out to GPTZero's team. GPTZero uncovers 50+ Hallucinations in ICLR 2026 GPTZero used our Hallucination Check tool to find 50+ hallucinations under review at ICLR, each of which were missed by 3-5 peer reviewers. [square_log]AI Detection Resources | GPTZeroPaul Esau [ICLR_Logo] The Deloitte Citation Situation - $98K Controversy Explained GPTZero used our Citation Check to analyze the 234 page report and identified more than 30 issues out of the total 141 citations, including 19 hallucinations. Using GPTZero's citation check would have saved ~$5000 per citation, all within minutes. [square_log]AI Detection Resources | GPTZeroNazar Shmatko [pexels-ani] Making America Hallucinate Again? GPTZero Detects New Errors in Major Government Report On May 22, the U.S. Presidential Commission to Make America Healthy Again (MAHA), led by health secretary Robert F. Kennedy Jr., released a major report on the causes of chronic diseases in children. Yet within a week, news outlets including NOTUS, the New York Times and Washington Post reported [square_log]AI Detection Resources | GPTZeroPaul Esau [Screenshot] * Investigations * Case Study * Research Written by Nazar Shmatko Keep reading Investigations Featured Paul Esau, Nazar Shmatko, Alex Adam Investigations Featured Paul Esau, Nazar Shmatko, Alex Adam GPTZero finds over 50 new hallucinations in ICLR 2026 submissions GPTZero used our Hallucination Check tool to find 50+ hallucinations under review at ICLR, each of which were missed by 3-5 peer reviewers. Dec 05, 2025 * 19 min read Investigations Featured Nazar Shmatko, Paul Esau, Alex Cui Investigations Featured Nazar Shmatko, Paul Esau, Alex Cui Deloitte's Citation Situation & GPTZero's Citation Solution GPTZero used our Citation Check to analyze the 234 page report and identified more than 30 issues out of the total 141 citations, including 19 hallucinations. Using GPTZero's citation check would have saved ~$5000 per citation, all within minutes. Nov 26, 2025 * 9 min read Share on X Copy link [solid-logo-2] Products * AI Detector * Chrome Extension * Integrations * Plagiarism Checker Resources * Pricing * Sales * Blog * Education Company * About us * Team * Affiliates