[{"date_created":"2025-02-05T07:49:00Z","quality_controlled":"1","date_updated":"2025-07-10T11:51:32Z","volume":37,"month":"12","_id":"19005","oa_version":"Published Version","scopus_import":"1","tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"has_accepted_license":"1","conference":{"end_date":"2024-12-16","location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems","start_date":"2024-12-16"},"type":"conference","related_material":{"link":[{"url":"https://github.com/CausalLearningAI/crl-dynamical-systems","relation":"software"}]},"date_published":"2024-12-01T00:00:00Z","citation":{"ama":"Yao D, Muller CJ, Locatello F. Marrying causal representation learning with dynamical systems for science. In: 38th Conference on Neural Information Processing Systems. Vol 37. Neural Information Processing Systems Foundation; 2024.","apa":"Yao, D., Muller, C. J., & Locatello, F. (2024). Marrying causal representation learning with dynamical systems for science. In 38th Conference on Neural Information Processing Systems (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ieee":"D. Yao, C. J. Muller, and F. Locatello, “Marrying causal representation learning with dynamical systems for science,” in 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024, vol. 37.","ista":"Yao D, Muller CJ, Locatello F. 2024. Marrying causal representation learning with dynamical systems for science. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","mla":"Yao, Dingling, et al. “Marrying Causal Representation Learning with Dynamical Systems for Science.” 38th Conference on Neural Information Processing Systems, vol. 37, Neural Information Processing Systems Foundation, 2024.","short":"D. Yao, C.J. Muller, F. Locatello, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","chicago":"Yao, Dingling, Caroline J Muller, and Francesco Locatello. “Marrying Causal Representation Learning with Dynamical Systems for Science.” In 38th Conference on Neural Information Processing Systems, Vol. 37. Neural Information Processing Systems Foundation, 2024."},"intvolume":" 37","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","corr_author":"1","status":"public","alternative_title":["Advances in Neural Information Processing Systems"],"arxiv":1,"department":[{"_id":"CaMu"},{"_id":"FrLo"}],"file_date_updated":"2025-02-05T07:44:58Z","publisher":"Neural Information Processing Systems Foundation","abstract":[{"text":"Causal representation learning promises to extend causal models to hidden causal\r\nvariables from raw entangled measurements. However, most progress has focused\r\non proving identifiability results in different settings, and we are not aware of any\r\nsuccessful real-world application. At the same time, the field of dynamical systems\r\nbenefited from deep learning and scaled to countless applications but does not allow\r\nparameter identification. In this paper, we draw a clear connection between the two\r\nand their key assumptions, allowing us to apply identifiable methods developed\r\nin causal representation learning to dynamical systems. At the same time, we can\r\nleverage scalable differentiable solvers developed for differential equations to build\r\nmodels that are both identifiable and practical. Overall, we learn explicitly controllable models that isolate the trajectory-specific parameters for further downstream\r\ntasks such as out-of-distribution classification or treatment effect estimation. We\r\nexperiment with a wind simulator with partially known factors of variation. We\r\nalso apply the resulting model to real-world climate data and successfully answer\r\ndownstream causal questions in line with existing literature on climate change.\r\nCode is available at https://github.com/CausalLearningAI/crl-dynamical-systems.","lang":"eng"}],"title":"Marrying causal representation learning with dynamical systems for science","publication_status":"published","acknowledgement":"We thank Niklas Boers for recommending the SpeedyWeather simulator and Valentino Maiorca\r\nfor guidance on Fourier transformation for SST data. We are also grateful to Shimeng Huang and Riccardo Cadei for their feedback on the treatment effect estimation experiment and to Jiale Chen and Adeel Pervez for their assistance with the solver implementation. Finally, we appreciate the anonymous reviewers for their insightful suggestions, which helped improve the manuscript. ","oa":1,"OA_place":"publisher","article_processing_charge":"No","OA_type":"gold","ddc":["000","550"],"day":"01","author":[{"last_name":"Yao","id":"d3e02e50-48a8-11ee-8f62-c108061797fa","full_name":"Yao, Dingling","first_name":"Dingling"},{"full_name":"Muller, Caroline J","first_name":"Caroline J","orcid":"0000-0001-5836-5350","last_name":"Muller","id":"f978ccb0-3f7f-11eb-b193-b0e2bd13182b"},{"first_name":"Francesco","full_name":"Locatello, Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","orcid":"0000-0002-4850-0683"}],"language":[{"iso":"eng"}],"external_id":{"arxiv":["2405.13888"]},"year":"2024","publication":"38th Conference on Neural Information Processing Systems","file":[{"checksum":"fe8832367e7143876f178244385d859e","relation":"main_file","file_id":"19006","file_name":"2024_NeurIPS_Yao.pdf","access_level":"open_access","content_type":"application/pdf","file_size":2595855,"date_created":"2025-02-05T07:44:58Z","date_updated":"2025-02-05T07:44:58Z","creator":"dernst","success":1}]},{"alternative_title":["Advances in Neural Information Processing Systems"],"arxiv":1,"status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","intvolume":" 37","citation":{"ama":"Kori A, Locatello F, Santhirasekaram A, Toni F, Glocker B, De Sousa Ribeiro F. Identifiable object-centric representation learning via probabilistic slot attention. In: 38th Conference on Neural Information Processing Systems. Vol 37. Neural Information Processing Systems Foundation; 2024.","ieee":"A. Kori, F. Locatello, A. Santhirasekaram, F. Toni, B. Glocker, and F. De Sousa Ribeiro, “Identifiable object-centric representation learning via probabilistic slot attention,” in 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024, vol. 37.","apa":"Kori, A., Locatello, F., Santhirasekaram, A., Toni, F., Glocker, B., & De Sousa Ribeiro, F. (2024). Identifiable object-centric representation learning via probabilistic slot attention. In 38th Conference on Neural Information Processing Systems (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ista":"Kori A, Locatello F, Santhirasekaram A, Toni F, Glocker B, De Sousa Ribeiro F. 2024. Identifiable object-centric representation learning via probabilistic slot attention. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","chicago":"Kori, Avinash, Francesco Locatello, Ainkaran Santhirasekaram, Francesca Toni, Ben Glocker, and Fabio De Sousa Ribeiro. “Identifiable Object-Centric Representation Learning via Probabilistic Slot Attention.” In 38th Conference on Neural Information Processing Systems, Vol. 37. Neural Information Processing Systems Foundation, 2024.","short":"A. Kori, F. Locatello, A. Santhirasekaram, F. Toni, B. Glocker, F. De Sousa Ribeiro, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","mla":"Kori, Avinash, et al. “Identifiable Object-Centric Representation Learning via Probabilistic Slot Attention.” 38th Conference on Neural Information Processing Systems, vol. 37, Neural Information Processing Systems Foundation, 2024."},"date_published":"2024-12-01T00:00:00Z","publisher":"Neural Information Processing Systems Foundation","department":[{"_id":"FrLo"}],"file_date_updated":"2025-02-05T08:34:25Z","scopus_import":"1","volume":37,"date_created":"2025-02-05T08:36:22Z","quality_controlled":"1","date_updated":"2025-05-14T11:29:10Z","_id":"19007","oa_version":"Published Version","month":"12","type":"conference","has_accepted_license":"1","conference":{"start_date":"2024-12-16","end_date":"2024-12-16","name":"NeurIPS: Neural Information Processing Systems","location":"Vancouver, Canada"},"year":"2024","language":[{"iso":"eng"}],"external_id":{"arxiv":["2406.07141"]},"file":[{"content_type":"application/pdf","access_level":"open_access","success":1,"creator":"dernst","file_size":6943800,"date_created":"2025-02-05T08:34:25Z","date_updated":"2025-02-05T08:34:25Z","checksum":"d27b3c7102adc28e798fe41001f0b919","file_name":"2024_NeurIPS_Kori.pdf","file_id":"19008","relation":"main_file"}],"publication":"38th Conference on Neural Information Processing Systems","acknowledgement":"A. Kori is supported by UKRI (grant number EP/S023356/1), as part of the UKRI Centre for Doctoral Training in Safe and Trusted AI. B. Glocker and F.D.S. Ribeiro acknowledge the support of the UKRI AI programme, and the Engineering and Physical Sciences Research Council, for CHAI - EPSRC Causality in Healthcare AI Hub (grant number EP/Y028856/1).","publication_status":"published","title":"Identifiable object-centric representation learning via probabilistic slot attention","abstract":[{"text":"Learning modular object-centric representations is crucial for systematic generalization. Existing methods show promising object-binding capabilities empirically,\r\nbut theoretical identifiability guarantees remain relatively underdeveloped. Understanding when object-centric representations can theoretically be identified is\r\ncrucial for scaling slot-based methods to high-dimensional images with correctness\r\nguarantees. To that end, we propose a probabilistic slot-attention algorithm that\r\nimposes an aggregate mixture prior over object-centric slot representations, thereby\r\nproviding slot identifiability guarantees without supervision, up to an equivalence\r\nrelation. We provide empirical verification of our theoretical identifiability result\r\nusing both simple 2-dimensional data and high-resolution imaging datasets.\r\n","lang":"eng"}],"day":"01","ddc":["000"],"author":[{"first_name":"Avinash","full_name":"Kori, Avinash","last_name":"Kori"},{"first_name":"Francesco","full_name":"Locatello, Francesco","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","orcid":"0000-0002-4850-0683"},{"full_name":"Santhirasekaram, Ainkaran","first_name":"Ainkaran","last_name":"Santhirasekaram"},{"last_name":"Toni","first_name":"Francesca","full_name":"Toni, Francesca"},{"full_name":"Glocker, Ben","first_name":"Ben","last_name":"Glocker"},{"full_name":"De Sousa Ribeiro, Fabio","first_name":"Fabio","last_name":"De Sousa Ribeiro"}],"article_processing_charge":"No","OA_type":"hybrid","OA_place":"publisher","oa":1},{"department":[{"_id":"TiBr"}],"publication":"arXiv","status":"public","corr_author":"1","year":"2024","arxiv":1,"citation":{"mla":"Glas, Jakob, and Matthew Hase-Liu. “Terminal Singularities of the Moduli Space of Curves on Low Degree Hypersurfaces and the Circle Method.” ArXiv, doi:10.48550/arXiv.2412.14923.","short":"J. Glas, M. Hase-Liu, ArXiv (n.d.).","chicago":"Glas, Jakob, and Matthew Hase-Liu. “Terminal Singularities of the Moduli Space of Curves on Low Degree Hypersurfaces and the Circle Method.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2412.14923.","ista":"Glas J, Hase-Liu M. Terminal singularities of the moduli space of curves on low degree hypersurfaces and the circle method. arXiv, 10.48550/arXiv.2412.14923.","apa":"Glas, J., & Hase-Liu, M. (n.d.). Terminal singularities of the moduli space of curves on low degree hypersurfaces and the circle method. arXiv. https://doi.org/10.48550/arXiv.2412.14923","ieee":"J. Glas and M. Hase-Liu, “Terminal singularities of the moduli space of curves on low degree hypersurfaces and the circle method,” arXiv. .","ama":"Glas J, Hase-Liu M. Terminal singularities of the moduli space of curves on low degree hypersurfaces and the circle method. arXiv. doi:10.48550/arXiv.2412.14923"},"date_published":"2024-12-19T00:00:00Z","language":[{"iso":"eng"}],"external_id":{"arxiv":["2412.14923"]},"user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","day":"19","related_material":{"record":[{"id":"18295","relation":"earlier_version","status":"public"}]},"author":[{"last_name":"Glas","id":"d6423cba-dc74-11ea-a0a7-ee61689ff5fb","first_name":"Jakob","full_name":"Glas, Jakob"},{"first_name":"Matthew ","full_name":"Hase-Liu, Matthew ","last_name":"Hase-Liu"}],"tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"OA_place":"repository","oa":1,"article_processing_charge":"No","type":"preprint","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2412.14923"}],"publication_status":"draft","doi":"10.48550/arXiv.2412.14923","abstract":[{"text":"We study the singularities of the moduli space of degree e maps from smooth genus g curves to an arbitrary smooth hypersurface of low degree. For e large compared to g, we show that these moduli spaces have at worst terminal singularities. Our main approach is to study the jet schemes of these moduli spaces by developing a suitable form of the circle method.","lang":"eng"}],"date_created":"2025-02-07T12:04:11Z","date_updated":"2025-04-15T08:05:40Z","_id":"19013","month":"12","oa_version":"Preprint","title":"Terminal singularities of the moduli space of curves on low degree hypersurfaces and the circle method"},{"date_published":"2024-07-01T00:00:00Z","citation":{"ama":"Browning TD. The polynomial sieve and equal sums of like polynomials. International Mathematics Research Notices. 2024;2024(13):10165-10168. doi:10.1093/imrn/rnae066","apa":"Browning, T. D. (2024). The polynomial sieve and equal sums of like polynomials. International Mathematics Research Notices. Oxford University Press. https://doi.org/10.1093/imrn/rnae066","ieee":"T. D. Browning, “The polynomial sieve and equal sums of like polynomials,” International Mathematics Research Notices, vol. 2024, no. 13. Oxford University Press, pp. 10165–10168, 2024.","ista":"Browning TD. 2024. The polynomial sieve and equal sums of like polynomials. International Mathematics Research Notices. 2024(13), 10165–10168.","mla":"Browning, Timothy D. “The Polynomial Sieve and Equal Sums of like Polynomials.” International Mathematics Research Notices, vol. 2024, no. 13, Oxford University Press, 2024, pp. 10165–68, doi:10.1093/imrn/rnae066.","chicago":"Browning, Timothy D. “The Polynomial Sieve and Equal Sums of like Polynomials.” International Mathematics Research Notices. Oxford University Press, 2024. https://doi.org/10.1093/imrn/rnae066.","short":"T.D. Browning, International Mathematics Research Notices 2024 (2024) 10165–10168."},"user_id":"317138e5-6ab7-11ef-aa6d-ffef3953e345","intvolume":" 2024","corr_author":"1","status":"public","department":[{"_id":"TiBr"}],"file_date_updated":"2025-02-18T07:56:36Z","publisher":"Oxford University Press","quality_controlled":"1","volume":2024,"date_updated":"2025-09-09T12:16:45Z","date_created":"2025-02-18T07:15:50Z","month":"07","_id":"19051","oa_version":"Published Version","scopus_import":"1","doi":"10.1093/imrn/rnae066","tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"has_accepted_license":"1","type":"journal_article","related_material":{"record":[{"id":"254","relation":"earlier_version","status":"public"}]},"language":[{"iso":"eng"}],"external_id":{"isi":["001196957300001"]},"article_type":"original","year":"2024","publication":"International Mathematics Research Notices","publication_identifier":{"eissn":["1687-0247"],"issn":["1073-7928"]},"isi":1,"file":[{"checksum":"b625b8adf018d2a97591813c1fc17b96","file_name":"2024_IMRN_Browning.pdf","relation":"main_file","file_id":"19052","access_level":"open_access","content_type":"application/pdf","creator":"dernst","success":1,"date_updated":"2025-02-18T07:56:36Z","file_size":205750,"date_created":"2025-02-18T07:56:36Z"}],"abstract":[{"text":"This paper corrects an error in an earlier work of the author.","lang":"eng"}],"title":"The polynomial sieve and equal sums of like polynomials","page":"10165-10168","publication_status":"published","OA_place":"publisher","oa":1,"article_processing_charge":"Yes (via OA deal)","OA_type":"hybrid","issue":"13","ddc":["510"],"day":"01","author":[{"last_name":"Browning","id":"35827D50-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0002-8314-0177","full_name":"Browning, Timothy D","first_name":"Timothy D"}]},{"oa_version":"Preprint","_id":"19063","month":"03","date_created":"2025-02-20T10:13:42Z","date_updated":"2025-02-24T12:52:23Z","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2403.06833","open_access":"1"}],"doi":"10.48550/arXiv.2403.06833","has_accepted_license":"1","tmp":{"image":"/images/cc_by_sa.png","name":"Creative Commons Attribution-ShareAlike 4.0 International Public License (CC BY-SA 4.0)","legal_code_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","short":"CC BY-SA (4.0)"},"type":"preprint","related_material":{"link":[{"relation":"software","url":" https://github.com/egozverev/Shold-It-Be-Executed-Or-Processed"}]},"date_published":"2024-03-01T00:00:00Z","citation":{"apa":"Zverev, E., Abdelnabi, S., Tabesh, S., Fritz, M., & Lampert, C. (2024). Can LLMs separate instructions from data? And what do we even mean by that? arXiv. https://doi.org/10.48550/arXiv.2403.06833","ieee":"E. Zverev, S. Abdelnabi, S. Tabesh, M. Fritz, and C. Lampert, “Can LLMs separate instructions from data? And what do we even mean by that?,” arXiv. 2024.","ama":"Zverev E, Abdelnabi S, Tabesh S, Fritz M, Lampert C. Can LLMs separate instructions from data? And what do we even mean by that? arXiv. 2024. doi:10.48550/arXiv.2403.06833","mla":"Zverev, Egor, et al. “Can LLMs Separate Instructions from Data? And What Do We Even Mean by That?” ArXiv, 2403.06833, 2024, doi:10.48550/arXiv.2403.06833.","short":"E. Zverev, S. Abdelnabi, S. Tabesh, M. Fritz, C. Lampert, ArXiv (2024).","chicago":"Zverev, Egor, Sahar Abdelnabi, Soroush Tabesh, Mario Fritz, and Christoph Lampert. “Can LLMs Separate Instructions from Data? And What Do We Even Mean by That?” ArXiv, 2024. https://doi.org/10.48550/arXiv.2403.06833.","ista":"Zverev E, Abdelnabi S, Tabesh S, Fritz M, Lampert C. 2024. Can LLMs separate instructions from data? And what do we even mean by that? arXiv, 2403.06833."},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","corr_author":"1","status":"public","arxiv":1,"article_number":"2403.06833","file_date_updated":"2025-02-20T10:11:45Z","department":[{"_id":"GradSch"},{"_id":"ChLa"}],"abstract":[{"text":"Instruction-tuned Large Language Models (LLMs) show impressive results in numerous practical applications, but they lack essential safety features that are common in other areas of computer science, particularly an explicit separation of instructions and data. This makes them vulnerable to manipulations such as indirect prompt injections and generally unsuitable for safety-critical tasks. Surprisingly, there is currently no established definition or benchmark to quantify this phenomenon. In this work, we close this gap by introducing a formal measure for instruction-data separation and an empirical variant that is calculable from a model's outputs. We also present a new dataset, SEP, that allows estimating the measure for real-world models. Our results on various LLMs show that the problem of instruction-data separation is real: all models fail to achieve high separation, and canonical mitigation techniques, such as prompt engineering and fine-tuning, either fail to substantially improve separation or reduce model utility. The source code and SEP dataset are openly accessible at https://github.com/egozverev/Shold-It-Be-Executed-Or-Processed.\r\n","lang":"eng"}],"title":"Can LLMs separate instructions from data? And what do we even mean by that?","publication_status":"published","acknowledgement":"The authors would like to sincerely thank Juan Rocamonde for valuable feedback to our manuscript. We acknowledge the support from the Scientific Service Units (SSU) of ISTA through resources provided by Scientific Computing (SciComp). We thank Dan Alistarh for providing us with computational resources. This work was partially funded by the German Federal Ministry of Education and Research (BMBF) under the grant AIgenCY (16KIS2012) and ELSA – European Lighthouse on Secure and Safe AI funded by the European Union under grant agreement No. 101070617. Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union or European Commission. Neither the European Union nor the European Commission can be held responsible for them.","oa":1,"OA_place":"repository","OA_type":"green","article_processing_charge":"No","author":[{"first_name":"Egor","full_name":"Zverev, Egor","id":"05162b19-1340-11ed-8f02-fa94e0e8c3bc","last_name":"Zverev"},{"full_name":"Abdelnabi, Sahar","first_name":"Sahar","last_name":"Abdelnabi"},{"orcid":"0009-0003-4119-6281","last_name":"Tabesh","id":"06000900-6068-11ef-8d61-c2472ef2e752","full_name":"Tabesh, Soroush","first_name":"Soroush"},{"last_name":"Fritz","first_name":"Mario","full_name":"Fritz, Mario"},{"orcid":"0000-0001-8622-7887","last_name":"Lampert","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","first_name":"Christoph","full_name":"Lampert, Christoph"}],"ddc":["000"],"day":"01","external_id":{"arxiv":["2403.06833"]},"language":[{"iso":"eng"}],"year":"2024","publication":"arXiv","file":[{"date_created":"2025-02-20T10:11:45Z","file_size":530972,"date_updated":"2025-02-20T10:11:45Z","success":1,"creator":"ezverev","access_level":"open_access","content_type":"application/pdf","relation":"main_file","file_id":"19064","file_name":"2403.06833v3.pdf","checksum":"35eb43968684b87be59144603ef10af0"}],"acknowledged_ssus":[{"_id":"ScienComp"}]},{"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","citation":{"ama":"Hwong Y-L, Muller CJ. Data - The unreasonable efficiency of total rain evaporation removal in triggering convective self-aggregation. 2024. doi:10.5281/ZENODO.10687169","ieee":"Y.-L. Hwong and C. J. Muller, “Data - The unreasonable efficiency of total rain evaporation removal in triggering convective self-aggregation.” Zenodo, 2024.","apa":"Hwong, Y.-L., & Muller, C. J. (2024). Data - The unreasonable efficiency of total rain evaporation removal in triggering convective self-aggregation. Zenodo. https://doi.org/10.5281/ZENODO.10687169","ista":"Hwong Y-L, Muller CJ. 2024. Data - The unreasonable efficiency of total rain evaporation removal in triggering convective self-aggregation, Zenodo, 10.5281/ZENODO.10687169.","short":"Y.-L. Hwong, C.J. Muller, (2024).","chicago":"Hwong, Yi-Ling, and Caroline J Muller. “Data - The Unreasonable Efficiency of Total Rain Evaporation Removal in Triggering Convective Self-Aggregation.” Zenodo, 2024. https://doi.org/10.5281/ZENODO.10687169.","mla":"Hwong, Yi-Ling, and Caroline J. Muller. Data - The Unreasonable Efficiency of Total Rain Evaporation Removal in Triggering Convective Self-Aggregation. Zenodo, 2024, doi:10.5281/ZENODO.10687169."},"date_published":"2024-02-21T00:00:00Z","year":"2024","corr_author":"1","status":"public","department":[{"_id":"CaMu"}],"publisher":"Zenodo","title":"Data - The unreasonable efficiency of total rain evaporation removal in triggering convective self-aggregation","_id":"19307","month":"02","oa_version":"Published Version","date_updated":"2025-09-04T13:16:39Z","date_created":"2025-03-07T08:39:40Z","abstract":[{"lang":"eng","text":"This repository contains the data, scripts, SAM codes and files required to reproduce the results of the manuscript \"The Unreasonable Efficiency of Total Rain Evaporation Removal in Triggering Convective Self-Aggregation\" submitted to the Geophysical Research Letters (GRL).\r\n\r\nBrief description of project: This project aims to examine the impact of rain evaporation removal or reduction in the planetary boundary layer (PBL) on convective self aggregation (CSA). Non-rotating radiative-convective equilibrium (RCE) simulations were conducted with the System for Atmospheric Modeling (SAM) cloud resolving model. Rain evaporation in the lowest 1 km was progressively reduced and the effect on CSA was investigated. The physical processes underlying this type of aggregation (referred to in the manuscript as no-evaporation CSA, or NE-CSA) were analyzed and described. \r\nThe default SAM code base (version 6.10.8) can be downloaded from here: http://rossby.msrc.sunysb.edu/~marat/SAM.html"}],"doi":"10.5281/ZENODO.10687169","main_file_link":[{"url":"https://doi.org/10.5281/zenodo.8369509","open_access":"1"}],"OA_type":"green","type":"research_data_reference","article_processing_charge":"No","tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"oa":1,"OA_place":"repository","has_accepted_license":"1","author":[{"orcid":"0000-0001-9281-3479","id":"1217aa61-4dd1-11ec-9ac3-f2ba3f17ee22","last_name":"Hwong","first_name":"Yi-Ling","full_name":"Hwong, Yi-Ling"},{"id":"f978ccb0-3f7f-11eb-b193-b0e2bd13182b","last_name":"Muller","orcid":"0000-0001-5836-5350","first_name":"Caroline J","full_name":"Muller, Caroline J"}],"related_material":{"record":[{"relation":"used_in_publication","id":"15186","status":"public"}]},"ddc":["550"],"day":"21"},{"publication_status":"published","title":"Continual learning: Applications and the road forward","abstract":[{"lang":"eng","text":"Continual learning is a subfield of machine learning, which aims to allow machine learning models to continuously learn on new data, by accumulating knowledge without forgetting what was learned in the past. In this work, we take a step back, and ask: \"Why should one care about continual learning in the first place?\". We set the stage by examining recent continual learning papers published at four major machine learning conferences, and show that memory-constrained settings dominate the field. Then, we discuss five open problems in machine learning, and even though they might seem unrelated to continual learning at first sight, we show that continual learning will inevitably be part of their solution. These problems are model editing, personalization and specialization, on-device learning, faster (re-)training and reinforcement learning. Finally, by comparing the desiderata from these unsolved problems and the current assumptions in continual learning, we highlight and discuss four future directions for continual learning research. We hope that this work offers an interesting perspective on the future of continual learning, while displaying its potential value and the paths we have to pursue in order to make it successful. This work is the result of the many discussions the authors had at the Dagstuhl seminar on Deep Continual Learning, in March 2023."}],"day":"12","ddc":["000"],"author":[{"first_name":"Eli","full_name":"Verwimp, Eli","last_name":"Verwimp"},{"last_name":"Aljundi","full_name":"Aljundi, Rahaf","first_name":"Rahaf"},{"first_name":"Shai","full_name":"Ben-David, Shai","last_name":"Ben-David"},{"last_name":"Bethge","first_name":"Matthias","full_name":"Bethge, Matthias"},{"last_name":"Cossu","first_name":"Andrea","full_name":"Cossu, Andrea"},{"last_name":"Gepperth","full_name":"Gepperth, Alexander","first_name":"Alexander"},{"last_name":"Hayes","first_name":"Tyler L.","full_name":"Hayes, Tyler L."},{"last_name":"Hüllermeier","full_name":"Hüllermeier, Eyke","first_name":"Eyke"},{"last_name":"Kanan","full_name":"Kanan, Christopher","first_name":"Christopher"},{"full_name":"Kudithipudi, Dhireesha","first_name":"Dhireesha","last_name":"Kudithipudi"},{"full_name":"Lampert, Christoph","first_name":"Christoph","id":"40C20FD2-F248-11E8-B48F-1D18A9856A87","last_name":"Lampert","orcid":"0000-0001-8622-7887"},{"last_name":"Mundt","full_name":"Mundt, Martin","first_name":"Martin"},{"last_name":"Pascanu","full_name":"Pascanu, Razvan","first_name":"Razvan"},{"last_name":"Popescu","first_name":"Adrian","full_name":"Popescu, Adrian"},{"first_name":"Andreas S.","full_name":"Tolias, Andreas S.","last_name":"Tolias"},{"last_name":"Van De Weijer","first_name":"Joost","full_name":"Van De Weijer, Joost"},{"last_name":"Liu","first_name":"Bing","full_name":"Liu, Bing"},{"last_name":"Lomonaco","first_name":"Vincenzo","full_name":"Lomonaco, Vincenzo"},{"first_name":"Tinne","full_name":"Tuytelaars, Tinne","last_name":"Tuytelaars"},{"last_name":"Van De Ven","first_name":"Gido M.","full_name":"Van De Ven, Gido M."}],"article_processing_charge":"No","OA_type":"diamond","oa":1,"OA_place":"publisher","article_type":"original","year":"2024","language":[{"iso":"eng"}],"external_id":{"arxiv":["2311.11908"]},"file":[{"checksum":"0714e12f7423cd098976ed9974561155","file_name":"2024_TMLR_Verwimp.pdf","relation":"main_file","file_id":"19426","access_level":"open_access","content_type":"application/pdf","creator":"dernst","success":1,"file_size":1367966,"date_updated":"2025-03-20T09:02:18Z","date_created":"2025-03-20T09:02:18Z"}],"publication_identifier":{"eissn":["2835-8856"]},"publication":"Transactions on Machine Learning Research","scopus_import":"1","quality_controlled":"1","date_updated":"2025-03-20T09:21:02Z","date_created":"2025-03-16T23:01:25Z","volume":2024,"oa_version":"Published Version","_id":"19408","month":"04","type":"journal_article","has_accepted_license":"1","tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"alternative_title":["TMLR"],"arxiv":1,"status":"public","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","intvolume":" 2024","date_published":"2024-04-12T00:00:00Z","citation":{"ieee":"E. Verwimp et al., “Continual learning: Applications and the road forward,” Transactions on Machine Learning Research, vol. 2024. Transactions on Machine Learning Research, 2024.","apa":"Verwimp, E., Aljundi, R., Ben-David, S., Bethge, M., Cossu, A., Gepperth, A., … Van De Ven, G. M. (2024). Continual learning: Applications and the road forward. Transactions on Machine Learning Research. Transactions on Machine Learning Research.","ama":"Verwimp E, Aljundi R, Ben-David S, et al. Continual learning: Applications and the road forward. Transactions on Machine Learning Research. 2024;2024.","chicago":"Verwimp, Eli, Rahaf Aljundi, Shai Ben-David, Matthias Bethge, Andrea Cossu, Alexander Gepperth, Tyler L. Hayes, et al. “Continual Learning: Applications and the Road Forward.” Transactions on Machine Learning Research. Transactions on Machine Learning Research, 2024.","short":"E. Verwimp, R. Aljundi, S. Ben-David, M. Bethge, A. Cossu, A. Gepperth, T.L. Hayes, E. Hüllermeier, C. Kanan, D. Kudithipudi, C. Lampert, M. Mundt, R. Pascanu, A. Popescu, A.S. Tolias, J. Van De Weijer, B. Liu, V. Lomonaco, T. Tuytelaars, G.M. Van De Ven, Transactions on Machine Learning Research 2024 (2024).","mla":"Verwimp, Eli, et al. “Continual Learning: Applications and the Road Forward.” Transactions on Machine Learning Research, vol. 2024, Transactions on Machine Learning Research, 2024.","ista":"Verwimp E, Aljundi R, Ben-David S, Bethge M, Cossu A, Gepperth A, Hayes TL, Hüllermeier E, Kanan C, Kudithipudi D, Lampert C, Mundt M, Pascanu R, Popescu A, Tolias AS, Van De Weijer J, Liu B, Lomonaco V, Tuytelaars T, Van De Ven GM. 2024. Continual learning: Applications and the road forward. Transactions on Machine Learning Research. 2024."},"publisher":"Transactions on Machine Learning Research","department":[{"_id":"ChLa"}],"file_date_updated":"2025-03-20T09:02:18Z"},{"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","citation":{"chicago":"Karle, Volker, Mikhail Lemeshko, Adrien Bouhon, Robert-Jan Slager, and F. Nur Ünal. “Anomalous Multi-Gap Topological Phases in Periodically Driven Quantum Rotors.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2408.16848.","short":"V. Karle, M. Lemeshko, A. Bouhon, R.-J. Slager, F.N. Ünal, ArXiv (n.d.).","mla":"Karle, Volker, et al. “Anomalous Multi-Gap Topological Phases in Periodically Driven Quantum Rotors.” ArXiv, 2408.16848, doi:10.48550/arXiv.2408.16848.","ista":"Karle V, Lemeshko M, Bouhon A, Slager R-J, Ünal FN. Anomalous multi-gap topological phases in periodically driven quantum rotors. arXiv, 2408.16848.","ieee":"V. Karle, M. Lemeshko, A. Bouhon, R.-J. Slager, and F. N. Ünal, “Anomalous multi-gap topological phases in periodically driven quantum rotors,” arXiv. .","apa":"Karle, V., Lemeshko, M., Bouhon, A., Slager, R.-J., & Ünal, F. N. (n.d.). Anomalous multi-gap topological phases in periodically driven quantum rotors. arXiv. https://doi.org/10.48550/arXiv.2408.16848","ama":"Karle V, Lemeshko M, Bouhon A, Slager R-J, Ünal FN. Anomalous multi-gap topological phases in periodically driven quantum rotors. arXiv. doi:10.48550/arXiv.2408.16848"},"date_published":"2024-08-29T00:00:00Z","article_number":"2408.16848","arxiv":1,"status":"public","ec_funded":1,"department":[{"_id":"MiLe"}],"project":[{"name":"Angulon: physics and applications of a new quasiparticle","_id":"2688CF98-B435-11E9-9278-68D0E5697425","grant_number":"801770","call_identifier":"H2020"}],"_id":"19425","month":"08","oa_version":"Preprint","date_created":"2025-03-20T07:48:23Z","date_updated":"2026-04-07T11:48:53Z","doi":"10.48550/arXiv.2408.16848","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2408.16848"}],"type":"preprint","related_material":{"record":[{"status":"public","relation":"dissertation_contains","id":"19393"}]},"external_id":{"arxiv":["2408.16848"]},"language":[{"iso":"eng"}],"year":"2024","publication":"arXiv","title":"Anomalous multi-gap topological phases in periodically driven quantum rotors","abstract":[{"lang":"eng","text":"We demonstrate that periodically driven quantum rotors provide a promising and broadly applicable platform to implement multi-gap topological phases, where groups of bands can acquire topological invariants due to non-Abelian braiding of band degeneracies. By adiabatically varying the periodic kicks to the rotor we find nodal-line braiding, which causes sign flips of topological charges of band nodes and can prevent them from annihilating, indicated by non-zero values of the %non-Abelian patch Euler class. In particular, we report\r\non the emergence of an anomalous Dirac string phase arising in the strongly driven regime, a truly out-of-equilibrium phase of the quantum rotor. This phase emanates from braiding processes involving all (quasienergy) gaps and manifests itself with edge states at zero angular momentum. Our results reveal direct applications in state-of-the-art experiments of quantum rotors, such as linear molecules driven by periodic far-off-resonant laser pulses or artificial\r\nquantum rotors in optical lattices, whose extensive versatility offers precise modification and observation of novel non-Abelian topological properties. "}],"acknowledgement":"We thank G. M. Koutentakis, S. Wimberger, J. G. E. Harris, T. Enss and A. Ghazaryan for fruitful discussions. M.L. acknowledges support by the European Research Council (ERC) Starting Grant No. 801770 (ANGULON). R.-J. S. acknowledges funding from a EPSRC ERC underwrite grant EP/X025829/1, a EPSRC New Investigator Award grant EP/W00187X/1, as well as Trinity College, Cambridge. F.N.U. acknowledges support from the Marie ¨Sk lodowska-Curie programme of the European Commission [Grant No. 893915], Simons Investigator Award\r\n[Grant No. 511029] and Trinity College Cambridge.","publication_status":"draft","OA_type":"green","article_processing_charge":"No","OA_place":"repository","oa":1,"author":[{"id":"D7C012AE-D7ED-11E9-95E8-1EC5E5697425","last_name":"Karle","orcid":"0000-0002-6963-0129","full_name":"Karle, Volker","first_name":"Volker"},{"full_name":"Lemeshko, Mikhail","first_name":"Mikhail","orcid":"0000-0002-6990-7802","id":"37CB05FA-F248-11E8-B48F-1D18A9856A87","last_name":"Lemeshko"},{"full_name":"Bouhon, Adrien","first_name":"Adrien","last_name":"Bouhon"},{"first_name":"Robert-Jan","full_name":"Slager, Robert-Jan","last_name":"Slager"},{"first_name":"F. Nur","full_name":"Ünal, F. Nur","last_name":"Ünal"}],"day":"29"},{"citation":{"ieee":"F. Nees et al., “Large-scale population data enrichment in mental health research,” Nature Mental Health, vol. 2, no. 10. Springer Nature, pp. 1124–1127, 2024.","apa":"Nees, F., Renner, P., Holz, N. E., Polemiti, E., Siehl, S., Hese, S., … Ogoh, G. (2024). Large-scale population data enrichment in mental health research. Nature Mental Health. Springer Nature. https://doi.org/10.1038/s44220-024-00316-z","ama":"Nees F, Renner P, Holz NE, et al. Large-scale population data enrichment in mental health research. Nature Mental Health. 2024;2(10):1124-1127. doi:10.1038/s44220-024-00316-z","chicago":"Nees, Frauke, Paul Renner, Nathalie E. Holz, Elli Polemiti, Sebastian Siehl, Sören Hese, Kerstin Schepanski, et al. “Large-Scale Population Data Enrichment in Mental Health Research.” Nature Mental Health. Springer Nature, 2024. https://doi.org/10.1038/s44220-024-00316-z.","short":"F. Nees, P. Renner, N.E. Holz, E. Polemiti, S. Siehl, S. Hese, K. Schepanski, G. Schumann, H. Walter, A. Heinz, M. Ralser, S. Twardziok, N. Vaidya, A. Bernas, E. Serin, M. Jentsch, E. Hitchen, H. Kebir, T.A. Lett, J.C. Roy, R. Eils, U.H. Taron, T. Schütz, J. Banks, T. Banaschewski, K. Jansone, N. Christmann, A. Meyer-Lindenberg, H. Tost, N. Holz, E. Schwarz, A. Stringaris, M. Neidhart, B. Seefried, R. Aden, O.A. Andreassen, L.T. Westlye, D. Van Der Meer, S. Fernandez, R. Kjelkenes, H. Ask, M. Rapp, M. Tschorn, S.J. Böttger, A. Marquand, G. Novarino, L. Marr, M. Slater, G.F. Viapiana, F.E. Orosa, J. Gallego, A. Pastor, A.J. Forstner, P. Hoffmann, M.M. Nöthen, I. Claus, A. Miller, C.M. Mathey, S. Heilmann-Heimbach, P. Sommer, M. Patraskaki, J. Wilbertz, K. Schmitt, V. Jirsa, S. Petkoski, S. Pitel, L. Otten, A.P. Athanasiadis, C. Pearmund, B. Spanlang, E. Alvarez, M. Sanchez, A. Giner, T. Jia, Y. Gong, Y. Xia, X. Chang, V. Calhoun, J. Liu, A. Schwalber, P. Thompson, N. Clinton, S. Desrivières, A.H. Young, B. Stahl, G. Ogoh, Nature Mental Health 2 (2024) 1124–1127.","mla":"Nees, Frauke, et al. “Large-Scale Population Data Enrichment in Mental Health Research.” Nature Mental Health, vol. 2, no. 10, Springer Nature, 2024, pp. 1124–27, doi:10.1038/s44220-024-00316-z.","ista":"Nees F, Renner P, Holz NE, Polemiti E, Siehl S, Hese S, Schepanski K, Schumann G, Walter H, Heinz A, Ralser M, Twardziok S, Vaidya N, Bernas A, Serin E, Jentsch M, Hitchen E, Kebir H, Lett TA, Roy JC, Eils R, Taron UH, Schütz T, Banks J, Banaschewski T, Jansone K, Christmann N, Meyer-Lindenberg A, Tost H, Holz N, Schwarz E, Stringaris A, Neidhart M, Seefried B, Aden R, Andreassen OA, Westlye LT, Van Der Meer D, Fernandez S, Kjelkenes R, Ask H, Rapp M, Tschorn M, Böttger SJ, Marquand A, Novarino G, Marr L, Slater M, Viapiana GF, Orosa FE, Gallego J, Pastor A, Forstner AJ, Hoffmann P, Nöthen MM, Claus I, Miller A, Mathey CM, Heilmann-Heimbach S, Sommer P, Patraskaki M, Wilbertz J, Schmitt K, Jirsa V, Petkoski S, Pitel S, Otten L, Athanasiadis AP, Pearmund C, Spanlang B, Alvarez E, Sanchez M, Giner A, Jia T, Gong Y, Xia Y, Chang X, Calhoun V, Liu J, Schwalber A, Thompson P, Clinton N, Desrivières S, Young AH, Stahl B, Ogoh G. 2024. Large-scale population data enrichment in mental health research. Nature Mental Health. 2(10), 1124–1127."},"date_published":"2024-10-01T00:00:00Z","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","intvolume":" 2","status":"public","department":[{"_id":"GaNo"}],"publisher":"Springer Nature","_id":"19446","month":"10","oa_version":"None","date_created":"2025-03-23T23:01:28Z","volume":2,"quality_controlled":"1","date_updated":"2025-03-25T08:28:39Z","scopus_import":"1","doi":"10.1038/s44220-024-00316-z","type":"journal_article","language":[{"iso":"eng"}],"year":"2024","article_type":"letter_note","publication":"Nature Mental Health","publication_identifier":{"eissn":["2731-6076"]},"abstract":[{"text":"This Comment explores new approaches to enrich large-scale population data, including incorporating macro-environmental and digital health measures.","lang":"eng"}],"title":"Large-scale population data enrichment in mental health research","publication_status":"published","page":"1124-1127","acknowledgement":"Funded by the European Union. Complementary funding was received by the UK Research and Innovation (UKRI) under the UK government’s Horizon Europe funding guarantee (10041392 and 10038599). Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union, the European Health and Digital Executive Agency (HADEA) or UKRI. The European Union, HADEA and UKRI cannot be held responsible for them. This work received also support from Chinese Ministry for Science and Technology (MOST), the Horizon 2020-funded European Research Council Advanced Grant ‘STRATIFY’ (695313), the German Research Foundation (COPE; 675346; NE 1383/15-1 (CoviDrug)) and the National Natural Science Foundation of China grant 82150710554.","OA_type":"closed access","article_processing_charge":"No","issue":"10","author":[{"full_name":"Nees, Frauke","first_name":"Frauke","last_name":"Nees"},{"first_name":"Paul","full_name":"Renner, Paul","last_name":"Renner"},{"full_name":"Holz, Nathalie E.","first_name":"Nathalie E.","last_name":"Holz"},{"last_name":"Polemiti","full_name":"Polemiti, Elli","first_name":"Elli"},{"last_name":"Siehl","first_name":"Sebastian","full_name":"Siehl, Sebastian"},{"last_name":"Hese","first_name":"Sören","full_name":"Hese, Sören"},{"full_name":"Schepanski, Kerstin","first_name":"Kerstin","last_name":"Schepanski"},{"full_name":"Schumann, Gunter","first_name":"Gunter","last_name":"Schumann"},{"first_name":"Henrik","full_name":"Walter, Henrik","last_name":"Walter"},{"last_name":"Heinz","full_name":"Heinz, Andreas","first_name":"Andreas"},{"full_name":"Ralser, Markus","first_name":"Markus","last_name":"Ralser"},{"first_name":"Sven","full_name":"Twardziok, Sven","last_name":"Twardziok"},{"last_name":"Vaidya","first_name":"Nilakshi","full_name":"Vaidya, Nilakshi"},{"last_name":"Bernas","first_name":"Antoine","full_name":"Bernas, Antoine"},{"last_name":"Serin","full_name":"Serin, Emin","first_name":"Emin"},{"full_name":"Jentsch, Marcel","first_name":"Marcel","last_name":"Jentsch"},{"first_name":"Esther","full_name":"Hitchen, Esther","last_name":"Hitchen"},{"first_name":"Hedi","full_name":"Kebir, Hedi","last_name":"Kebir"},{"full_name":"Lett, Tristram A.","first_name":"Tristram A.","last_name":"Lett"},{"first_name":"Jean Charles","full_name":"Roy, Jean Charles","last_name":"Roy"},{"last_name":"Eils","first_name":"Roland","full_name":"Eils, Roland"},{"full_name":"Taron, Ulrike Helene","first_name":"Ulrike Helene","last_name":"Taron"},{"last_name":"Schütz","full_name":"Schütz, Tatjana","first_name":"Tatjana"},{"first_name":"Jamie","full_name":"Banks, Jamie","last_name":"Banks"},{"full_name":"Banaschewski, Tobias","first_name":"Tobias","last_name":"Banaschewski"},{"full_name":"Jansone, Karina","first_name":"Karina","last_name":"Jansone"},{"full_name":"Christmann, Nina","first_name":"Nina","last_name":"Christmann"},{"first_name":"Andreas","full_name":"Meyer-Lindenberg, Andreas","last_name":"Meyer-Lindenberg"},{"last_name":"Tost","first_name":"Heike","full_name":"Tost, Heike"},{"last_name":"Holz","full_name":"Holz, Nathalie","first_name":"Nathalie"},{"last_name":"Schwarz","first_name":"Emanuel","full_name":"Schwarz, Emanuel"},{"last_name":"Stringaris","full_name":"Stringaris, Argyris","first_name":"Argyris"},{"last_name":"Neidhart","first_name":"Maja","full_name":"Neidhart, Maja"},{"full_name":"Seefried, Beke","first_name":"Beke","last_name":"Seefried"},{"first_name":"Rieke","full_name":"Aden, Rieke","last_name":"Aden"},{"last_name":"Andreassen","full_name":"Andreassen, Ole A.","first_name":"Ole A."},{"last_name":"Westlye","first_name":"Lars T.","full_name":"Westlye, Lars T."},{"first_name":"Dennis","full_name":"Van Der Meer, Dennis","last_name":"Van Der Meer"},{"first_name":"Sara","full_name":"Fernandez, Sara","last_name":"Fernandez"},{"full_name":"Kjelkenes, Rikka","first_name":"Rikka","last_name":"Kjelkenes"},{"last_name":"Ask","first_name":"Helga","full_name":"Ask, Helga"},{"first_name":"Michael","full_name":"Rapp, Michael","last_name":"Rapp"},{"last_name":"Tschorn","full_name":"Tschorn, Mira","first_name":"Mira"},{"full_name":"Böttger, Sarah Jane","first_name":"Sarah Jane","last_name":"Böttger"},{"last_name":"Marquand","full_name":"Marquand, Andre","first_name":"Andre"},{"orcid":"0000-0002-7673-7178","last_name":"Novarino","id":"3E57A680-F248-11E8-B48F-1D18A9856A87","first_name":"Gaia","full_name":"Novarino, Gaia"},{"first_name":"Lena","full_name":"Marr, Lena","id":"4406F586-F248-11E8-B48F-1D18A9856A87","last_name":"Marr"},{"last_name":"Slater","full_name":"Slater, Mel","first_name":"Mel"},{"last_name":"Viapiana","full_name":"Viapiana, Guillem Feixas","first_name":"Guillem Feixas"},{"last_name":"Orosa","first_name":"Francisco Eiroa","full_name":"Orosa, Francisco Eiroa"},{"first_name":"Jaime","full_name":"Gallego, Jaime","last_name":"Gallego"},{"last_name":"Pastor","first_name":"Alvaro","full_name":"Pastor, Alvaro"},{"last_name":"Forstner","full_name":"Forstner, Andreas J.","first_name":"Andreas J."},{"last_name":"Hoffmann","first_name":"Per","full_name":"Hoffmann, Per"},{"last_name":"Nöthen","first_name":"Markus M.","full_name":"Nöthen, Markus M."},{"full_name":"Claus, Isabelle","first_name":"Isabelle","last_name":"Claus"},{"full_name":"Miller, Abigail","first_name":"Abigail","last_name":"Miller"},{"first_name":"Carina M.","full_name":"Mathey, Carina M.","last_name":"Mathey"},{"last_name":"Heilmann-Heimbach","full_name":"Heilmann-Heimbach, Stefanie","first_name":"Stefanie"},{"last_name":"Sommer","full_name":"Sommer, Peter","first_name":"Peter"},{"last_name":"Patraskaki","first_name":"Myrto","full_name":"Patraskaki, Myrto"},{"last_name":"Wilbertz","full_name":"Wilbertz, Johannes","first_name":"Johannes"},{"last_name":"Schmitt","full_name":"Schmitt, Karen","first_name":"Karen"},{"last_name":"Jirsa","first_name":"Viktor","full_name":"Jirsa, Viktor"},{"first_name":"Spase","full_name":"Petkoski, Spase","last_name":"Petkoski"},{"first_name":"Séverine","full_name":"Pitel, Séverine","last_name":"Pitel"},{"full_name":"Otten, Lisa","first_name":"Lisa","last_name":"Otten"},{"full_name":"Athanasiadis, Anastasios Polykarpos","first_name":"Anastasios Polykarpos","last_name":"Athanasiadis"},{"full_name":"Pearmund, Charlie","first_name":"Charlie","last_name":"Pearmund"},{"first_name":"Bernhard","full_name":"Spanlang, Bernhard","last_name":"Spanlang"},{"last_name":"Alvarez","full_name":"Alvarez, Elena","first_name":"Elena"},{"first_name":"Mavi","full_name":"Sanchez, Mavi","last_name":"Sanchez"},{"first_name":"Arantxa","full_name":"Giner, Arantxa","last_name":"Giner"},{"last_name":"Jia","full_name":"Jia, Tianye","first_name":"Tianye"},{"last_name":"Gong","full_name":"Gong, Yanting","first_name":"Yanting"},{"first_name":"Yunman","full_name":"Xia, Yunman","last_name":"Xia"},{"full_name":"Chang, Xiao","first_name":"Xiao","last_name":"Chang"},{"full_name":"Calhoun, Vince","first_name":"Vince","last_name":"Calhoun"},{"full_name":"Liu, Jingyu","first_name":"Jingyu","last_name":"Liu"},{"last_name":"Schwalber","first_name":"Ameli","full_name":"Schwalber, Ameli"},{"full_name":"Thompson, Paul","first_name":"Paul","last_name":"Thompson"},{"first_name":"Nicholas","full_name":"Clinton, Nicholas","last_name":"Clinton"},{"last_name":"Desrivières","first_name":"Sylvane","full_name":"Desrivières, Sylvane"},{"last_name":"Young","first_name":"Allan H.","full_name":"Young, Allan H."},{"last_name":"Stahl","first_name":"Bernd","full_name":"Stahl, Bernd"},{"last_name":"Ogoh","first_name":"George","full_name":"Ogoh, George"}],"day":"01"},{"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","intvolume":" 37","date_published":"2024-12-20T00:00:00Z","citation":{"ista":"Modoranu I-V, Safaryan M, Malinovsky G, Kurtic E, Robert T, Richtárik P, Alistarh D-A. 2024. MICROADAM: Accurate adaptive optimization with low space overhead and provable convergence. 38th Conference on Neural Information Processing Systems. , Advances in Neural Information Processing Systems, vol. 37.","mla":"Modoranu, Ionut-Vlad, et al. “MICROADAM: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence.” 38th Conference on Neural Information Processing Systems, vol. 37, Neural Information Processing Systems Foundation, 2024.","short":"I.-V. Modoranu, M. Safaryan, G. Malinovsky, E. Kurtic, T. Robert, P. Richtárik, D.-A. Alistarh, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","chicago":"Modoranu, Ionut-Vlad, Mher Safaryan, Grigory Malinovsky, Eldar Kurtic, Thomas Robert, Peter Richtárik, and Dan-Adrian Alistarh. “MICROADAM: Accurate Adaptive Optimization with Low Space Overhead and Provable Convergence.” In 38th Conference on Neural Information Processing Systems, Vol. 37. Neural Information Processing Systems Foundation, 2024.","ama":"Modoranu I-V, Safaryan M, Malinovsky G, et al. MICROADAM: Accurate adaptive optimization with low space overhead and provable convergence. In: 38th Conference on Neural Information Processing Systems. Vol 37. Neural Information Processing Systems Foundation; 2024.","apa":"Modoranu, I.-V., Safaryan, M., Malinovsky, G., Kurtic, E., Robert, T., Richtárik, P., & Alistarh, D.-A. (2024). MICROADAM: Accurate adaptive optimization with low space overhead and provable convergence. In 38th Conference on Neural Information Processing Systems (Vol. 37). Neural Information Processing Systems Foundation.","ieee":"I.-V. Modoranu et al., “MICROADAM: Accurate adaptive optimization with low space overhead and provable convergence,” in 38th Conference on Neural Information Processing Systems, 2024, vol. 37."},"arxiv":1,"alternative_title":["Advances in Neural Information Processing Systems"],"corr_author":"1","status":"public","ec_funded":1,"department":[{"_id":"DaAl"}],"publisher":"Neural Information Processing Systems Foundation","scopus_import":"1","project":[{"name":"IST-BRIDGE: International postdoctoral program","_id":"fc2ed2f7-9c52-11eb-aca3-c01059dda49c","grant_number":"101034413","call_identifier":"H2020"}],"_id":"19510","month":"12","oa_version":"Preprint","quality_controlled":"1","date_created":"2025-04-06T22:01:32Z","volume":37,"date_updated":"2025-05-14T11:32:52Z","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2405.15593"}],"type":"conference","related_material":{"link":[{"url":"https://github.com/IST-DASLab/MicroAdam","relation":"software"}]},"external_id":{"arxiv":["2405.15593"]},"language":[{"iso":"eng"}],"year":"2024","publication":"38th Conference on Neural Information Processing Systems","acknowledged_ssus":[{"_id":"CampIT"}],"publication_identifier":{"issn":["1049-5258"]},"title":"MICROADAM: Accurate adaptive optimization with low space overhead and provable convergence","abstract":[{"text":"We propose a new variant of the Adam optimizer [Kingma and Ba, 2014] called\r\nMICROADAM that specifically minimizes memory overheads, while maintaining\r\ntheoretical convergence guarantees. We achieve this by compressing the gradient\r\ninformation before it is fed into the optimizer state, thereby reducing its memory\r\nfootprint significantly. We control the resulting compression error via a novel\r\ninstance of the classical error feedback mechanism from distributed optimization [Seide et al., 2014, Alistarh et al., 2018, Karimireddy et al., 2019] in which\r\nthe error correction information is itself compressed to allow for practical memory\r\ngains. We prove that the resulting approach maintains theoretical convergence\r\nguarantees competitive to those of AMSGrad, while providing good practical performance. Specifically, we show that MICROADAM can be implemented efficiently\r\non GPUs: on both million-scale (BERT) and billion-scale (LLaMA) models, MICROADAM provides practical convergence competitive to that of the uncompressed\r\nAdam baseline, with lower memory usage and similar running time. Our code is\r\navailable at https://github.com/IST-DASLab/MicroAdam.","lang":"eng"}],"acknowledgement":"The authors thank Razvan Pascanu, Mahdi Nikdan and Soroush Tabesh for their valuable feedback, the IT department from Institute of Science and Technology Austria for the hardware support and Weights and Biases for the infrastructure to track all our experiments. Mher Safaryan has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement No 101034413.","publication_status":"published","OA_type":"green","article_processing_charge":"No","oa":1,"OA_place":"repository","author":[{"id":"449f7a18-f128-11eb-9611-9b430c0c6333","last_name":"Modoranu","first_name":"Ionut-Vlad","full_name":"Modoranu, Ionut-Vlad"},{"last_name":"Safaryan","id":"dd546b39-0804-11ed-9c55-ef075c39778d","full_name":"Safaryan, Mher","first_name":"Mher"},{"last_name":"Malinovsky","first_name":"Grigory","full_name":"Malinovsky, Grigory"},{"full_name":"Kurtic, Eldar","first_name":"Eldar","last_name":"Kurtic","id":"47beb3a5-07b5-11eb-9b87-b108ec578218"},{"first_name":"Thomas","full_name":"Robert, Thomas","id":"de632733-1457-11f0-ae22-b5914b8c1c41","last_name":"Robert"},{"last_name":"Richtárik","first_name":"Peter","full_name":"Richtárik, Peter"},{"full_name":"Alistarh, Dan-Adrian","first_name":"Dan-Adrian","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","last_name":"Alistarh","orcid":"0000-0003-3650-940X"}],"day":"20"},{"publication_status":"published","abstract":[{"lang":"eng","text":"We introduce QuaRot, a new Quantization scheme based on Rotations, which is able to quantize LLMs end-to-end, including all weights, activations, and KV cache in 4 bits. QuaRot rotates LLMs in a way that removes outliers from the hidden state without changing the output, making quantization easier. This computational invariance is applied to the hidden state (residual) of the LLM, as well as to the activations of the feed-forward components, aspects of the attention mechanism, and to the KV cache. The result is a quantized model where all matrix multiplications are performed in 4 bits, without any channels identified for retention in higher precision. Our 4-bit quantized LLAMA2-70B model has losses of at most 0.47 WikiText-2 perplexity and retains 99% of the zero-shot performance. We also show that QuaRot can provide lossless 6 and 8 bit LLAMA-2 models without any calibration data using round-to-nearest quantization. Code is available at github.com/spcl/QuaRot."}],"title":"QuaRot: Outlier-free 4-bit inference in rotated LLMs","author":[{"full_name":"Ashkboos, Saleh","first_name":"Saleh","last_name":"Ashkboos"},{"last_name":"Mohtashami","full_name":"Mohtashami, Amirkeivan","first_name":"Amirkeivan"},{"last_name":"Croci","full_name":"Croci, Maximilian L.","first_name":"Maximilian L."},{"first_name":"Bo","full_name":"Li, Bo","last_name":"Li"},{"last_name":"Cameron","first_name":"Pashmina","full_name":"Cameron, Pashmina"},{"full_name":"Jaggi, Martin","first_name":"Martin","last_name":"Jaggi"},{"id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","last_name":"Alistarh","orcid":"0000-0003-3650-940X","full_name":"Alistarh, Dan-Adrian","first_name":"Dan-Adrian"},{"full_name":"Hoefler, Torsten","first_name":"Torsten","last_name":"Hoefler"},{"last_name":"Hensman","first_name":"James","full_name":"Hensman, James"}],"day":"20","OA_place":"repository","oa":1,"OA_type":"green","article_processing_charge":"No","year":"2024","external_id":{"arxiv":["2404.00456"]},"language":[{"iso":"eng"}],"publication_identifier":{"issn":["1049-5258"]},"publication":"38th Conference on Neural Information Processing Systems","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2404.00456","open_access":"1"}],"oa_version":"Preprint","_id":"19511","month":"12","quality_controlled":"1","date_updated":"2025-05-14T11:33:12Z","volume":37,"date_created":"2025-04-06T22:01:32Z","scopus_import":"1","related_material":{"link":[{"relation":"software","url":"https://github.com/spcl/QuaRot"}]},"conference":{"start_date":"2024-12-09","location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems","end_date":"2024-12-15"},"type":"conference","status":"public","arxiv":1,"alternative_title":["Advances in Neural Information Processing Systems"],"citation":{"chicago":"Ashkboos, Saleh, Amirkeivan Mohtashami, Maximilian L. Croci, Bo Li, Pashmina Cameron, Martin Jaggi, Dan-Adrian Alistarh, Torsten Hoefler, and James Hensman. “QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs.” In 38th Conference on Neural Information Processing Systems, Vol. 37. Neural Information Processing Systems Foundation, 2024.","short":"S. Ashkboos, A. Mohtashami, M.L. Croci, B. Li, P. Cameron, M. Jaggi, D.-A. Alistarh, T. Hoefler, J. Hensman, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","mla":"Ashkboos, Saleh, et al. “QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs.” 38th Conference on Neural Information Processing Systems, vol. 37, Neural Information Processing Systems Foundation, 2024.","ista":"Ashkboos S, Mohtashami A, Croci ML, Li B, Cameron P, Jaggi M, Alistarh D-A, Hoefler T, Hensman J. 2024. QuaRot: Outlier-free 4-bit inference in rotated LLMs. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","ieee":"S. Ashkboos et al., “QuaRot: Outlier-free 4-bit inference in rotated LLMs,” in 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024, vol. 37.","apa":"Ashkboos, S., Mohtashami, A., Croci, M. L., Li, B., Cameron, P., Jaggi, M., … Hensman, J. (2024). QuaRot: Outlier-free 4-bit inference in rotated LLMs. In 38th Conference on Neural Information Processing Systems (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ama":"Ashkboos S, Mohtashami A, Croci ML, et al. QuaRot: Outlier-free 4-bit inference in rotated LLMs. In: 38th Conference on Neural Information Processing Systems. Vol 37. Neural Information Processing Systems Foundation; 2024."},"date_published":"2024-12-20T00:00:00Z","intvolume":" 37","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publisher":"Neural Information Processing Systems Foundation","department":[{"_id":"DaAl"}]},{"department":[{"_id":"MoHe"}],"publisher":"Neural Information Processing Systems Foundation","date_published":"2024-12-20T00:00:00Z","citation":{"ista":"Andersson JD, Henzinger M, Pagh R, Steiner TA, Upadhyay J. 2024. Continual counting with gradual privacy expiration. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","mla":"Andersson, Joel Daniel, et al. “Continual Counting with Gradual Privacy Expiration.” 38th Conference on Neural Information Processing Systems, vol. 37, Neural Information Processing Systems Foundation, 2024.","chicago":"Andersson, Joel Daniel, Monika Henzinger, Rasmus Pagh, Teresa Anna Steiner, and Jalaj Upadhyay. “Continual Counting with Gradual Privacy Expiration.” In 38th Conference on Neural Information Processing Systems, Vol. 37. Neural Information Processing Systems Foundation, 2024.","short":"J.D. Andersson, M. Henzinger, R. Pagh, T.A. Steiner, J. Upadhyay, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","ama":"Andersson JD, Henzinger M, Pagh R, Steiner TA, Upadhyay J. Continual counting with gradual privacy expiration. In: 38th Conference on Neural Information Processing Systems. Vol 37. Neural Information Processing Systems Foundation; 2024.","apa":"Andersson, J. D., Henzinger, M., Pagh, R., Steiner, T. A., & Upadhyay, J. (2024). Continual counting with gradual privacy expiration. In 38th Conference on Neural Information Processing Systems (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ieee":"J. D. Andersson, M. Henzinger, R. Pagh, T. A. Steiner, and J. Upadhyay, “Continual counting with gradual privacy expiration,” in 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024, vol. 37."},"intvolume":" 37","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","ec_funded":1,"corr_author":"1","status":"public","alternative_title":["Advances in Neural Information Processing Systems"],"arxiv":1,"conference":{"start_date":"2024-12-09","end_date":"2024-12-15","location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems"},"type":"conference","quality_controlled":"1","volume":37,"date_created":"2025-04-06T22:01:32Z","date_updated":"2025-05-14T11:33:22Z","_id":"19512","month":"12","oa_version":"Preprint","project":[{"grant_number":"101019564","call_identifier":"H2020","name":"The design and evaluation of modern fully dynamic data structures","_id":"bd9ca328-d553-11ed-ba76-dc4f890cfe62"},{"grant_number":"Z00422","name":"Efficient algorithms","_id":"34def286-11ca-11ed-8bc3-da5948e1613c"},{"grant_number":"I05982","name":"Static and Dynamic Hierarchical Graph Decompositions","_id":"bda196b2-d553-11ed-ba76-8e8ee6c21103"},{"grant_number":"P33775","_id":"bd9e3a2e-d553-11ed-ba76-8aa684ce17fe","name":"Fast Algorithms for a Reactive Network Layer"}],"scopus_import":"1","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2406.03802"}],"publication":"38th Conference on Neural Information Processing Systems","publication_identifier":{"issn":["1049-5258"]},"language":[{"iso":"eng"}],"external_id":{"arxiv":["2406.03802"]},"year":"2024","oa":1,"OA_place":"repository","article_processing_charge":"No","OA_type":"green","day":"20","author":[{"first_name":"Joel Daniel","full_name":"Andersson, Joel Daniel","last_name":"Andersson"},{"last_name":"Henzinger","id":"540c9bbd-f2de-11ec-812d-d04a5be85630","orcid":"0000-0002-5008-6530","full_name":"Henzinger, Monika H","first_name":"Monika H"},{"last_name":"Pagh","first_name":"Rasmus","full_name":"Pagh, Rasmus"},{"last_name":"Steiner","first_name":"Teresa Anna","full_name":"Steiner, Teresa Anna"},{"first_name":"Jalaj","full_name":"Upadhyay, Jalaj","last_name":"Upadhyay"}],"abstract":[{"text":"Differential privacy with gradual expiration models the setting where data items\r\narrive in a stream and at a given time t the privacy loss guaranteed for a data item\r\nseen at time (t − d) is εg(d), where g is a monotonically non-decreasing function.\r\nWe study the fundamental continual (binary) counting problem where each data\r\nitem consists of a bit, and the algorithm needs to output at each time step the sum of\r\nall the bits streamed so far. For a stream of length T and privacy without expiration\r\ncontinual counting is possible with maximum (over all time steps) additive error\r\nO(log2\r\n(T)/ε) and the best known lower bound is Ω(log(T)/ε); closing this gap\r\nis a challenging open problem.\r\nWe show that the situation is very different for privacy with gradual expiration by\r\ngiving upper and lower bounds for a large set of expiration functions g. Specifically,\r\nour algorithm achieves an additive error of O(log(T)/ε) for a large set of privacy\r\nexpiration functions. We also give a lower bound that shows that if C is the additive\r\nerror of any ε-DP algorithm for this problem, then the product of C and the privacy\r\nexpiration function after 2C steps must be Ω(log(T)/ε). Our algorithm matches\r\nthis lower bound as its additive error is O(log(T)/ε), even when g(2C) = O(1).\r\nOur empirical evaluation shows that we achieve a slowly growing privacy loss\r\nwith significantly smaller empirical privacy loss for large values of d than a natural\r\nbaseline algorithm.","lang":"eng"}],"title":"Continual counting with gradual privacy expiration","publication_status":"published","acknowledgement":"Monika Henzinger: This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant agreement No. 101019564) and the Austrian Science Fund (FWF) grant DOI 10.55776/Z422, grant DOI 10.55776/I5982, and grant DOI 10.55776/P33775 with additional funding from the netidee SCIENCE Stiftung, 2020–2024. Joel Daniel Andersson and Rasmus Pagh are affiliated with Basic Algorithms Research Copenhagen (BARC), supported by the VILLUM Foundation grant 16582, and are also supported by Providentia, a Data Science Distinguished Investigator grant from Novo Nordisk Fonden. Teresa Anna Steiner is supported by a research grant (VIL51463) from VILLUM FONDEN. This work was done while Teresa Anna Steiner was a Postdoc at the Technical University of Denmark. Jalaj Upadhyay’s research was funded by the Rutgers Decanal Grant no. 302918 and an unrestricted gift from Google."},{"author":[{"first_name":"Marco","full_name":"Fumero, Marco","last_name":"Fumero","id":"1c1593eb-393f-11ef-bb8e-ab4f1e979650"},{"last_name":"Pegoraro","first_name":"Marco","full_name":"Pegoraro, Marco"},{"first_name":"Valentino","full_name":"Maiorca, Valentino","last_name":"Maiorca"},{"first_name":"Francesco","full_name":"Locatello, Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","orcid":"0000-0002-4850-0683"},{"first_name":"Emanuele","full_name":"Rodolà, Emanuele","last_name":"Rodolà"}],"day":"20","OA_type":"green","article_processing_charge":"No","oa":1,"OA_place":"repository","acknowledgement":"MF is supported by the MSCA IST-Bridge fellowship which has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 101034413. ER and VM are supported by the PNRR MUR project PE0000013-FAIR. MP is supported by the Sapienza grant \"Predicting and Explaining Clinical Trial Outcomes\", prot. RG12218166FA3F13.","publication_status":"published","title":"Latent functional maps: A spectral framework for representation alignment","abstract":[{"lang":"eng","text":"Neural models learn data representations that lie on low-dimensional manifolds,\r\nyet modeling the relation between these representational spaces is an ongoing challenge. By integrating spectral geometry principles into neural modeling, we show\r\nthat this problem can be better addressed in the functional domain, mitigating complexity, while enhancing interpretability and performances on downstream tasks.\r\nTo this end, we introduce a multi-purpose framework to the representation learning\r\ncommunity, which allows to: (i) compare different spaces in an interpretable way\r\nand measure their intrinsic similarity; (ii) find correspondences between them, both\r\nin unsupervised and weakly supervised settings, and (iii) to effectively transfer\r\nrepresentations between distinct spaces. We validate our framework on various\r\napplications, ranging from stitching to retrieval tasks, and on multiple modalities,\r\ndemonstrating that Latent Functional Maps can serve as a swiss-army knife for\r\nrepresentation alignment"}],"publication_identifier":{"issn":["1049-5258"]},"publication":"38th Conference on Neural Information Processing Systems","year":"2024","external_id":{"arxiv":["2406.14183"]},"language":[{"iso":"eng"}],"type":"conference","conference":{"start_date":"2024-12-09","name":"NeurIPS: Neural Information Processing Systems","location":"Vancouver, Canada","end_date":"2024-12-15"},"main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2406.14183"}],"scopus_import":"1","project":[{"name":"IST-BRIDGE: International postdoctoral program","_id":"fc2ed2f7-9c52-11eb-aca3-c01059dda49c","grant_number":"101034413","call_identifier":"H2020"}],"month":"12","_id":"19515","oa_version":"Preprint","quality_controlled":"1","date_updated":"2025-05-14T11:36:51Z","date_created":"2025-04-06T22:01:32Z","volume":37,"publisher":"Neural Information Processing Systems Foundation","department":[{"_id":"FrLo"}],"arxiv":1,"alternative_title":["Advances in Neural Information Processing Systems"],"corr_author":"1","status":"public","ec_funded":1,"intvolume":" 37","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","date_published":"2024-12-20T00:00:00Z","citation":{"ista":"Fumero M, Pegoraro M, Maiorca V, Locatello F, Rodolà E. 2024. Latent functional maps: A spectral framework for representation alignment. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","mla":"Fumero, Marco, et al. “Latent Functional Maps: A Spectral Framework for Representation Alignment.” 38th Conference on Neural Information Processing Systems, vol. 37, Neural Information Processing Systems Foundation, 2024.","chicago":"Fumero, Marco, Marco Pegoraro, Valentino Maiorca, Francesco Locatello, and Emanuele Rodolà. “Latent Functional Maps: A Spectral Framework for Representation Alignment.” In 38th Conference on Neural Information Processing Systems, Vol. 37. Neural Information Processing Systems Foundation, 2024.","short":"M. Fumero, M. Pegoraro, V. Maiorca, F. Locatello, E. Rodolà, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","ama":"Fumero M, Pegoraro M, Maiorca V, Locatello F, Rodolà E. Latent functional maps: A spectral framework for representation alignment. In: 38th Conference on Neural Information Processing Systems. Vol 37. Neural Information Processing Systems Foundation; 2024.","apa":"Fumero, M., Pegoraro, M., Maiorca, V., Locatello, F., & Rodolà, E. (2024). Latent functional maps: A spectral framework for representation alignment. In 38th Conference on Neural Information Processing Systems (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ieee":"M. Fumero, M. Pegoraro, V. Maiorca, F. Locatello, and E. Rodolà, “Latent functional maps: A spectral framework for representation alignment,” in 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024, vol. 37."}},{"publication":"38th Conference on Neural Information Processing Systems","publication_identifier":{"issn":["1049-5258"]},"language":[{"iso":"eng"}],"external_id":{"arxiv":["2405.17897"]},"year":"2024","OA_place":"repository","oa":1,"article_processing_charge":"No","OA_type":"green","day":"20","author":[{"last_name":"Crisostomi","full_name":"Crisostomi, Donato","first_name":"Donato"},{"first_name":"Marco","full_name":"Fumero, Marco","last_name":"Fumero","id":"1c1593eb-393f-11ef-bb8e-ab4f1e979650"},{"first_name":"Daniele","full_name":"Baieri, Daniele","last_name":"Baieri"},{"last_name":"Bernard","first_name":"Florian","full_name":"Bernard, Florian"},{"first_name":"Emanuele","full_name":"Rodolà, Emanuele","last_name":"Rodolà"}],"abstract":[{"lang":"eng","text":"In this paper, we present a novel data-free method for merging neural networks in weight space. Differently from most existing works, our method optimizes for the permutations of network neurons globally across all layers. This allows us to enforce cycle consistency of the permutations when merging n ≥ 3 models, allowing circular compositions of permutations to be computed without accumulating error along the path. We qualitatively and quantitatively motivate the need for such a constraint, showing its benefits when merging sets of models in scenarios spanning varying architectures and datasets. We finally show that, when coupled\r\nwith activation renormalization, our approach yields the best results in the task."}],"title":"C2M3: Cycle-consistent multi-model merging","publication_status":"published","acknowledgement":"This work is supported by the ERC grant no.802554 (SPECGEO), PRIN 2020 project\r\nno.2020TA3K9N (LEGO.AI), and PNRR MUR project PE0000013-FAIR. Marco Fumero is supported by the MSCA IST-Bridge fellowship which has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 101034413. We thank Simone Scardapane for the helpful feedback on the paper.","department":[{"_id":"FrLo"}],"publisher":"Neural Information Processing Systems Foundation","date_published":"2024-12-20T00:00:00Z","citation":{"ista":"Crisostomi D, Fumero M, Baieri D, Bernard F, Rodolà E. 2024. C2M3: Cycle-consistent multi-model merging. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","mla":"Crisostomi, Donato, et al. “C2M3: Cycle-Consistent Multi-Model Merging.” 38th Conference on Neural Information Processing Systems, vol. 37, Neural Information Processing Systems Foundation, 2024.","short":"D. Crisostomi, M. Fumero, D. Baieri, F. Bernard, E. Rodolà, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","chicago":"Crisostomi, Donato, Marco Fumero, Daniele Baieri, Florian Bernard, and Emanuele Rodolà. “C2M3: Cycle-Consistent Multi-Model Merging.” In 38th Conference on Neural Information Processing Systems, Vol. 37. Neural Information Processing Systems Foundation, 2024.","ama":"Crisostomi D, Fumero M, Baieri D, Bernard F, Rodolà E. C2M3: Cycle-consistent multi-model merging. In: 38th Conference on Neural Information Processing Systems. Vol 37. Neural Information Processing Systems Foundation; 2024.","apa":"Crisostomi, D., Fumero, M., Baieri, D., Bernard, F., & Rodolà, E. (2024). C2M3: Cycle-consistent multi-model merging. In 38th Conference on Neural Information Processing Systems (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ieee":"D. Crisostomi, M. Fumero, D. Baieri, F. Bernard, and E. Rodolà, “C2M3: Cycle-consistent multi-model merging,” in 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024, vol. 37."},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","intvolume":" 37","ec_funded":1,"corr_author":"1","status":"public","alternative_title":["Advances in Neural Information Processing Systems"],"arxiv":1,"conference":{"end_date":"2024-12-15","location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems","start_date":"2024-12-09"},"type":"conference","date_updated":"2025-05-14T11:36:59Z","volume":37,"quality_controlled":"1","date_created":"2025-04-06T22:01:32Z","month":"12","_id":"19517","oa_version":"Preprint","project":[{"_id":"fc2ed2f7-9c52-11eb-aca3-c01059dda49c","name":"IST-BRIDGE: International postdoctoral program","call_identifier":"H2020","grant_number":"101034413"}],"scopus_import":"1","main_file_link":[{"open_access":"1","url":"https://doi.org/10.48550/arXiv.2405.17897"}]},{"date_published":"2024-12-20T00:00:00Z","citation":{"short":"D. Wu, I.-V. Modoranu, M. Safaryan, D. Kuznedelev, D.-A. Alistarh, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","chicago":"Wu, Diyuan, Ionut-Vlad Modoranu, Mher Safaryan, Denis Kuznedelev, and Dan-Adrian Alistarh. “The Iterative Optimal Brain Surgeon: Faster Sparse Recovery by Leveraging Second-Order Information.” In 38th Conference on Neural Information Processing Systems, Vol. 37. Neural Information Processing Systems Foundation, 2024.","mla":"Wu, Diyuan, et al. “The Iterative Optimal Brain Surgeon: Faster Sparse Recovery by Leveraging Second-Order Information.” 38th Conference on Neural Information Processing Systems, vol. 37, Neural Information Processing Systems Foundation, 2024.","ista":"Wu D, Modoranu I-V, Safaryan M, Kuznedelev D, Alistarh D-A. 2024. The iterative optimal brain surgeon: Faster sparse recovery by leveraging second-order information. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","ieee":"D. Wu, I.-V. Modoranu, M. Safaryan, D. Kuznedelev, and D.-A. Alistarh, “The iterative optimal brain surgeon: Faster sparse recovery by leveraging second-order information,” in 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024, vol. 37.","apa":"Wu, D., Modoranu, I.-V., Safaryan, M., Kuznedelev, D., & Alistarh, D.-A. (2024). The iterative optimal brain surgeon: Faster sparse recovery by leveraging second-order information. In 38th Conference on Neural Information Processing Systems (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ama":"Wu D, Modoranu I-V, Safaryan M, Kuznedelev D, Alistarh D-A. The iterative optimal brain surgeon: Faster sparse recovery by leveraging second-order information. In: 38th Conference on Neural Information Processing Systems. Vol 37. Neural Information Processing Systems Foundation; 2024."},"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","intvolume":" 37","corr_author":"1","status":"public","ec_funded":1,"arxiv":1,"alternative_title":["Advances in Neural Information Processing Systems"],"department":[{"_id":"DaAl"},{"_id":"MaMo"}],"publisher":"Neural Information Processing Systems Foundation","month":"12","_id":"19518","oa_version":"Preprint","quality_controlled":"1","volume":37,"date_created":"2025-04-06T22:01:32Z","date_updated":"2025-05-14T11:37:10Z","scopus_import":"1","project":[{"_id":"fc2ed2f7-9c52-11eb-aca3-c01059dda49c","name":"IST-BRIDGE: International postdoctoral program","call_identifier":"H2020","grant_number":"101034413"}],"main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2408.17163","open_access":"1"}],"conference":{"location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems","end_date":"2024-12-15","start_date":"2024-12-09"},"type":"conference","external_id":{"arxiv":["2408.17163"]},"language":[{"iso":"eng"}],"year":"2024","publication":"38th Conference on Neural Information Processing Systems","publication_identifier":{"issn":["1049-5258"]},"acknowledged_ssus":[{"_id":"CampIT"}],"abstract":[{"text":"The rising footprint of machine learning has led to a focus on imposing model\r\nsparsity as a means of reducing computational and memory costs. For deep neural\r\nnetworks (DNNs), the state-of-the-art accuracy-vs-sparsity is achieved by heuristics\r\ninspired by the classical Optimal Brain Surgeon (OBS) framework [LeCun et al.,\r\n1989, Hassibi and Stork, 1992, Hassibi et al., 1993], which leverages loss curvature\r\ninformation to make better pruning decisions. Yet, these results still lack a solid\r\ntheoretical understanding, and it is unclear whether they can be improved by\r\nleveraging connections to the wealth of work on sparse recovery algorithms. In this\r\npaper, we draw new connections between these two areas and present new sparse\r\nrecovery algorithms inspired by the OBS framework that comes with theoretical\r\nguarantees under reasonable assumptions and have strong practical performance.\r\nSpecifically, our work starts from the observation that we can leverage curvature\r\ninformation in OBS-like fashion upon the projection step of classic iterative sparse\r\nrecovery algorithms such as IHT. We show for the first time that this leads both\r\nto improved convergence bounds under standard assumptions. Furthermore, we\r\npresent extensions of this approach to the practical task of obtaining accurate sparse\r\nDNNs, and validate it experimentally at scale for Transformer-based models on\r\nvision and language tasks.","lang":"eng"}],"title":"The iterative optimal brain surgeon: Faster sparse recovery by leveraging second-order information","publication_status":"published","acknowledgement":"The authors thank the anonymous NeurIPS reviewers for their useful comments and feedback, the IT department from the Institute of Science and Technology Austria for the hardware support, and Weights and Biases for the infrastructure to track all our experiments. Mher Safaryan has received funding from the European Union’s Horizon 2020 research and innovation program under the Maria Skłodowska-Curie grant agreement No 101034413.","oa":1,"OA_place":"repository","OA_type":"green","article_processing_charge":"No","author":[{"last_name":"Wu","id":"1a5914c2-896a-11ed-bdf8-fb80621a0635","full_name":"Wu, Diyuan","first_name":"Diyuan"},{"id":"449f7a18-f128-11eb-9611-9b430c0c6333","last_name":"Modoranu","first_name":"Ionut-Vlad","full_name":"Modoranu, Ionut-Vlad"},{"last_name":"Safaryan","id":"dd546b39-0804-11ed-9c55-ef075c39778d","full_name":"Safaryan, Mher","first_name":"Mher"},{"full_name":"Kuznedelev, Denis","first_name":"Denis","last_name":"Kuznedelev"},{"full_name":"Alistarh, Dan-Adrian","first_name":"Dan-Adrian","orcid":"0000-0003-3650-940X","last_name":"Alistarh","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87"}],"day":"20"},{"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","intvolume":" 37","citation":{"ama":"Malinovskii V, Mazur D, Ilin I, et al. PV-tuning: Beyond straight-through estimation for extreme LLM compression. In: 38th Conference on Neural Information Processing Systems. Vol 37. Neural Information Processing Systems Foundation; 2024.","ieee":"V. Malinovskii et al., “PV-tuning: Beyond straight-through estimation for extreme LLM compression,” in 38th Conference on Neural Information Processing Systems, Vancouver, Canada, 2024, vol. 37.","apa":"Malinovskii, V., Mazur, D., Ilin, I., Kuznedelev, D., Burlachenko, K., Yi, K., … Richtarik, P. (2024). PV-tuning: Beyond straight-through estimation for extreme LLM compression. In 38th Conference on Neural Information Processing Systems (Vol. 37). Vancouver, Canada: Neural Information Processing Systems Foundation.","ista":"Malinovskii V, Mazur D, Ilin I, Kuznedelev D, Burlachenko K, Yi K, Alistarh D-A, Richtarik P. 2024. PV-tuning: Beyond straight-through estimation for extreme LLM compression. 38th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 37.","chicago":"Malinovskii, Vladimir, Denis Mazur, Ivan Ilin, Denis Kuznedelev, Konstantin Burlachenko, Kai Yi, Dan-Adrian Alistarh, and Peter Richtarik. “PV-Tuning: Beyond Straight-through Estimation for Extreme LLM Compression.” In 38th Conference on Neural Information Processing Systems, Vol. 37. Neural Information Processing Systems Foundation, 2024.","short":"V. Malinovskii, D. Mazur, I. Ilin, D. Kuznedelev, K. Burlachenko, K. Yi, D.-A. Alistarh, P. Richtarik, in:, 38th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2024.","mla":"Malinovskii, Vladimir, et al. “PV-Tuning: Beyond Straight-through Estimation for Extreme LLM Compression.” 38th Conference on Neural Information Processing Systems, vol. 37, Neural Information Processing Systems Foundation, 2024."},"date_published":"2024-12-20T00:00:00Z","arxiv":1,"alternative_title":["Advances in Neural Information Processing Systems"],"status":"public","file_date_updated":"2025-04-07T09:17:10Z","department":[{"_id":"DaAl"}],"publisher":"Neural Information Processing Systems Foundation","scopus_import":"1","oa_version":"Published Version","_id":"19519","month":"12","date_updated":"2025-05-14T10:49:20Z","volume":37,"date_created":"2025-04-06T22:01:32Z","quality_controlled":"1","type":"conference","conference":{"end_date":"2024-12-15","location":"Vancouver, Canada","name":"NeurIPS: Neural Information Processing Systems","start_date":"2024-12-10"},"has_accepted_license":"1","external_id":{"arxiv":["2405.14852"]},"language":[{"iso":"eng"}],"year":"2024","publication":"38th Conference on Neural Information Processing Systems","file":[{"access_level":"open_access","content_type":"application/pdf","date_created":"2025-04-07T09:17:10Z","file_size":939712,"date_updated":"2025-04-07T09:17:10Z","success":1,"creator":"dernst","checksum":"54d36f947887e26d0e568b512167001a","relation":"main_file","file_id":"19521","file_name":"2024_NeurIPS_Malinovskii.pdf"}],"publication_identifier":{"isbn":["9798331314385"],"issn":["1049-5258"]},"title":"PV-tuning: Beyond straight-through estimation for extreme LLM compression","abstract":[{"lang":"eng","text":"There has been significant interest in \"extreme\" compression of large language models (LLMs), i.e. to 1-2 bits per parameter, which allows such models to be executed efficiently on resource-constrained devices. Existing work focused on improved one-shot quantization techniques and weight representations; yet, purely post-training approaches are reaching diminishing returns in terms of the accuracy-vs-bit-width trade-off. State-of-the-art quantization methods such as QuIP# and AQLM include fine-tuning (part of) the compressed parameters over a limited amount of calibration data; however, such fine-tuning techniques over compressed weights often make exclusive use of straight-through estimators (STE), whose performance is not well-understood in this setting. In this work, we question the use of STE for extreme LLM compression, showing that it can be sub-optimal, and perform a systematic study of quantization-aware fine-tuning strategies for LLMs.We propose PV-Tuning - a representation-agnostic framework that generalizes and improves upon existing fine-tuning strategies, and provides convergence guarantees in restricted cases.On the practical side, when used for 1-2 bit vector quantization, PV-Tuning outperforms prior techniques for highly-performant models such as Llama and Mistral. Using PV-Tuning, we achieve the first Pareto-optimal quantization for Llama-2 family models at 2 bits per parameter."}],"acknowledgement":"Authors would like to thank Vage Egiazarian, Andrei Panferov and Ruslan Svirschevski for their\r\nhelp and advice on AQLM codebase and running large-scale experiments. We also thank Philip\r\nZmushko and Artem Fedorov for helpful discussions during the early stages of our research. The research of Kai Yi, Konstantin Burlachenko, and Peter Richtárik reported in this publication was supported by funding from King Abdullah University of Science and Technology (KAUST) – Center of Excellence for Generative AI, under award number 5940. We would also like to thank our NeurIPS reviewers for their helpful suggestions, we specifically highlight p3Lv’s suggestions to consider smaller codebook sizes and evaluate PV-Tuning with QuIP#, both of which produced interesting findings. Finally, we thank the open-source contributors from llama.cpp9 and the LocalLlama10 community for discussions and inspirations on practical use cases of quantized language models, and in particular, Yalda Shabanzadeh and Arthur Aardvark for their help with improving the codebase.","publication_status":"published","OA_type":"gold","article_processing_charge":"No","oa":1,"OA_place":"publisher","author":[{"full_name":"Malinovskii, Vladimir","first_name":"Vladimir","last_name":"Malinovskii"},{"full_name":"Mazur, Denis","first_name":"Denis","last_name":"Mazur"},{"last_name":"Ilin","full_name":"Ilin, Ivan","first_name":"Ivan"},{"last_name":"Kuznedelev","full_name":"Kuznedelev, Denis","first_name":"Denis"},{"last_name":"Burlachenko","full_name":"Burlachenko, Konstantin","first_name":"Konstantin"},{"last_name":"Yi","first_name":"Kai","full_name":"Yi, Kai"},{"last_name":"Alistarh","id":"4A899BFC-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0003-3650-940X","first_name":"Dan-Adrian","full_name":"Alistarh, Dan-Adrian"},{"last_name":"Richtarik","first_name":"Peter","full_name":"Richtarik, Peter"}],"day":"20","ddc":["000"]},{"citation":{"apa":"Vijatovic, D., Toma, F. A., Harrington, Z. P., Sommer, C. M., Hauschild, R., Trevisan, A. J., … Sweeney, L. B. (n.d.). Spinal neuron diversity scales exponentially with swim-to-limb transformation during frog metamorphosis. bioRxiv. https://doi.org/10.1101/2024.09.20.614050","ieee":"D. Vijatovic et al., “Spinal neuron diversity scales exponentially with swim-to-limb transformation during frog metamorphosis,” bioRxiv. .","ama":"Vijatovic D, Toma FA, Harrington ZP, et al. Spinal neuron diversity scales exponentially with swim-to-limb transformation during frog metamorphosis. bioRxiv. doi:10.1101/2024.09.20.614050","mla":"Vijatovic, David, et al. “Spinal Neuron Diversity Scales Exponentially with Swim-to-Limb Transformation during Frog Metamorphosis.” BioRxiv, doi:10.1101/2024.09.20.614050.","short":"D. Vijatovic, F.A. Toma, Z.P. Harrington, C.M. Sommer, R. Hauschild, A.J. Trevisan, P. Chapman, M. Julseth, S. Brenner-Morton, M.I. Gabitto, J.S. Dasen, J.B. Bikoff, L.B. Sweeney, BioRxiv (n.d.).","chicago":"Vijatovic, David, Florina Alexandra Toma, Zoe P Harrington, Christoph M Sommer, Robert Hauschild, Alexandra J. Trevisan, Phillip Chapman, et al. “Spinal Neuron Diversity Scales Exponentially with Swim-to-Limb Transformation during Frog Metamorphosis.” BioRxiv, n.d. https://doi.org/10.1101/2024.09.20.614050.","ista":"Vijatovic D, Toma FA, Harrington ZP, Sommer CM, Hauschild R, Trevisan AJ, Chapman P, Julseth M, Brenner-Morton S, Gabitto MI, Dasen JS, Bikoff JB, Sweeney LB. Spinal neuron diversity scales exponentially with swim-to-limb transformation during frog metamorphosis. bioRxiv, 10.1101/2024.09.20.614050."},"date_published":"2024-09-27T00:00:00Z","language":[{"iso":"eng"}],"user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","corr_author":"1","status":"public","year":"2024","department":[{"_id":"LoSw"},{"_id":"TiVo"},{"_id":"Bio"},{"_id":"NiBa"}],"publication":"bioRxiv","acknowledged_ssus":[{"_id":"Bio"}],"date_updated":"2025-05-14T11:40:13Z","date_created":"2025-04-07T08:48:28Z","abstract":[{"lang":"eng","text":"Vertebrates exhibit a wide range of motor behaviors, ranging from swimming to complex limb-based movements. Here we take advantage of frog metamorphosis, which captures a swim-to-limb-based movement transformation during the development of a single organism, to explore changes in the underlying spinal circuits. We find that the tadpole spinal cord contains small and largely homogeneous populations of motor neurons (MNs) and V1 interneurons (V1s) at early escape swimming stages. These neuronal populations only modestly increase in number and subtype heterogeneity with the emergence of free swimming. In contrast, during frog metamorphosis and the emergence of limb movement, there is a dramatic expansion of MN and V1 interneuron number and transcriptional heterogeneity, culminating in cohorts of neurons that exhibit striking molecular similarity to mammalian motor circuits. CRISPR/Cas9-mediated gene disruption of the limb MN and V1 determinants FoxP1 and Engrailed-1, respectively, results in severe but selective deficits in tail and limb function. Our work thus demonstrates that neural diversity scales exponentially with increasing behavioral complexity and illustrates striking evolutionary conservation in the molecular organization and function of motor circuits across species."}],"oa_version":"Preprint","_id":"19520","month":"09","project":[{"name":"Development of V1 interneuron diversity during swim-to-walk transition of Xenopus metamorphosis","_id":"bd73af52-d553-11ed-ba76-912049f0ac7a","grant_number":"FTI21-D-046"},{"grant_number":"101041551","_id":"ebb66355-77a9-11ec-83b8-b8ac210a4dae","name":"Development and Evolution of Tetrapod Motor Circuits"},{"grant_number":"CZI01","_id":"c08e9ad1-5a5b-11eb-8a69-9d1cf3b07473","name":"Tools for automation and feedback microscopy"}],"title":"Spinal neuron diversity scales exponentially with swim-to-limb transformation during frog metamorphosis","main_file_link":[{"url":"https://doi.org/10.1101/2024.09.20.614050","open_access":"1"}],"publication_status":"submitted","acknowledgement":"We would like to thank the members of the Sweeney Lab (especially Stavros Papadopoulos and\r\nSophie Gobeil) for their contributions to this project and, in addition to the lab, Graziana Gatto\r\nand Mario de Bono, for discussion, and support. We are also grateful to Tom Jessell and Chris\r\nKintner for their scientific insight and mentorship during the conception of this project. This\r\nproject would also not have been possible with the technical support of the Matthias Nowak,\r\nVerena Mayer and the Aquatics as well as the Imaging and Optics Facility support teams\r\n(ISTA). In addition, we thank our funding sources for providing the resources to do these\r\nexperiments: FTI Strategy Lower Austria Dissertation Grant Number FT121-D-046 (D.V.);\r\nHorizon Europe ERC Starting Grant Number 101041551 (L.B.S., F.A.T. and D.V); Special\r\nResearch Program (SFB) of the Austrian Science Fund (FWF) Project number F7814-B (L.B.S);\r\nNINDS 5R35NS116858 (J.S.D); CZI grant DAF2020-225401 (DOI): 10.37921/120055ratwvi\r\n(R.H.); NIH grant number R01NS123116 (J.B.B); American Lebanese Syrian Associated\r\nCharities (ALSAC) (J.B.B.); German Academic Exchange Service (DAAD) IFI Grant Number\r\n57515251-91853472 (Z.H.); and Project A.L.S. (S.B-M.). ","doi":"10.1101/2024.09.20.614050","oa":1,"OA_place":"repository","article_processing_charge":"No","type":"preprint","OA_type":"green","day":"27","author":[{"last_name":"Vijatovic","id":"cf391e77-ec3c-11ea-a124-d69323410b58","full_name":"Vijatovic, David","first_name":"David"},{"id":"2f73f876-f128-11eb-9611-b96b5a30cb0e","last_name":"Toma","full_name":"Toma, Florina Alexandra ","first_name":"Florina Alexandra "},{"last_name":"Harrington","id":"a8144562-32c9-11ee-b5ce-d9800628bda2","orcid":"0009-0008-0158-4032","full_name":"Harrington, Zoe P","first_name":"Zoe P"},{"full_name":"Sommer, Christoph M","first_name":"Christoph M","orcid":"0000-0003-1216-9105","last_name":"Sommer","id":"4DF26D8C-F248-11E8-B48F-1D18A9856A87"},{"full_name":"Hauschild, Robert","first_name":"Robert","last_name":"Hauschild","id":"4E01D6B4-F248-11E8-B48F-1D18A9856A87","orcid":"0000-0001-9843-3522"},{"last_name":"Trevisan","first_name":"Alexandra J.","full_name":"Trevisan, Alexandra J."},{"full_name":"Chapman, Phillip","first_name":"Phillip","last_name":"Chapman"},{"id":"1cf464b2-dc7d-11ea-9b2f-f9b1aa9417d1","last_name":"Julseth","full_name":"Julseth, Mara","first_name":"Mara"},{"last_name":"Brenner-Morton","full_name":"Brenner-Morton, Susan","first_name":"Susan"},{"first_name":"Mariano I.","full_name":"Gabitto, Mariano I.","last_name":"Gabitto"},{"last_name":"Dasen","first_name":"Jeremy S.","full_name":"Dasen, Jeremy S."},{"first_name":"Jay B.","full_name":"Bikoff, Jay B.","last_name":"Bikoff"},{"full_name":"Sweeney, Lora Beatrice Jaeger","first_name":"Lora Beatrice Jaeger","orcid":"0000-0001-9242-5601","last_name":"Sweeney","id":"56BE8254-C4F0-11E9-8E45-0B23E6697425"}]},{"title":"Eigenstate thermalisation at the edge for Wigner matrices","abstract":[{"text":"We prove the Eigenstate Thermalisation Hypothesis for Wigner matrices\r\nuniformly in the entire spectrum, in particular near the spectral edges, with a\r\nbound on the fluctuation that is optimal for any observable. This complements\r\nearlier works of Cipolloni et. al. (Comm. Math. Phys. 388, 2021; Forum Math.,\r\nSigma 10, 2022) and Benigni et. al. (Comm. Math. Phys. 391, 2022; arXiv:\r\n2303.11142) that were restricted either to the bulk of the spectrum or to\r\nspecial observables. As a main ingredient, we prove a new multi-resolvent local\r\nlaw that optimally accounts for the edge scaling.","lang":"eng"}],"acknowledgement":"Supported by ERC Advanced Grant “RMTBeyond” No. 101020331.","publication_status":"draft","article_processing_charge":"No","OA_place":"repository","oa":1,"author":[{"id":"42198EFA-F248-11E8-B48F-1D18A9856A87","last_name":"Cipolloni","orcid":"0000-0002-4901-7992","full_name":"Cipolloni, Giorgio","first_name":"Giorgio"},{"id":"4DBD5372-F248-11E8-B48F-1D18A9856A87","last_name":"Erdös","orcid":"0000-0001-5366-9603","first_name":"László","full_name":"Erdös, László"},{"full_name":"Henheik, Sven Joscha","first_name":"Sven Joscha","orcid":"0000-0003-1106-327X","last_name":"Henheik","id":"31d731d7-d235-11ea-ad11-b50331c8d7fb"}],"day":"17","external_id":{"arxiv":["2309.05488"]},"language":[{"iso":"eng"}],"year":"2024","publication":"arXiv","project":[{"_id":"62796744-2b32-11ec-9570-940b20777f1d","name":"Random matrices beyond Wigner-Dyson-Mehta","call_identifier":"H2020","grant_number":"101020331"}],"oa_version":"Preprint","_id":"19545","month":"12","date_created":"2025-04-11T08:19:22Z","date_updated":"2026-04-07T12:37:11Z","doi":"10.48550/arXiv.2309.05488","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2309.05488","open_access":"1"}],"type":"preprint","tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"related_material":{"record":[{"status":"public","relation":"dissertation_contains","id":"19540"}]},"user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","date_published":"2024-12-17T00:00:00Z","citation":{"ama":"Cipolloni G, Erdös L, Henheik SJ. Eigenstate thermalisation at the edge for Wigner matrices. arXiv. doi:10.48550/arXiv.2309.05488","ieee":"G. Cipolloni, L. Erdös, and S. J. Henheik, “Eigenstate thermalisation at the edge for Wigner matrices,” arXiv. .","apa":"Cipolloni, G., Erdös, L., & Henheik, S. J. (n.d.). Eigenstate thermalisation at the edge for Wigner matrices. arXiv. https://doi.org/10.48550/arXiv.2309.05488","ista":"Cipolloni G, Erdös L, Henheik SJ. Eigenstate thermalisation at the edge for Wigner matrices. arXiv, 10.48550/arXiv.2309.05488.","short":"G. Cipolloni, L. Erdös, S.J. Henheik, ArXiv (n.d.).","chicago":"Cipolloni, Giorgio, László Erdös, and Sven Joscha Henheik. “Eigenstate Thermalisation at the Edge for Wigner Matrices.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2309.05488.","mla":"Cipolloni, Giorgio, et al. “Eigenstate Thermalisation at the Edge for Wigner Matrices.” ArXiv, doi:10.48550/arXiv.2309.05488."},"arxiv":1,"corr_author":"1","status":"public","ec_funded":1,"department":[{"_id":"LaEr"}]},{"day":"03","author":[{"first_name":"László","full_name":"Erdös, László","id":"4DBD5372-F248-11E8-B48F-1D18A9856A87","last_name":"Erdös","orcid":"0000-0001-5366-9603"},{"last_name":"Henheik","id":"31d731d7-d235-11ea-ad11-b50331c8d7fb","orcid":"0000-0003-1106-327X","full_name":"Henheik, Sven Joscha","first_name":"Sven Joscha"},{"first_name":"Volodymyr","full_name":"Riabov, Volodymyr","last_name":"Riabov","id":"1949f904-edfb-11eb-afb5-e2dfddabb93b"}],"article_processing_charge":"No","oa":1,"OA_place":"repository","acknowledgement":"Supported by the ERC Advanced Grant \"RMTBeyond\"\r\nNo. 101020331.","publication_status":"draft","title":"Cusp universality for correlated random matrices","abstract":[{"lang":"eng","text":"For correlated real symmetric or complex Hermitian random matrices, we prove\r\nthat the local eigenvalue statistics at any cusp singularity are universal.\r\nSince the density of states typically exhibits only square root edge or cubic\r\nroot cusp singularities, our result completes the proof of the\r\nWigner-Dyson-Mehta universality conjecture in all spectral regimes for a very\r\ngeneral class of random matrices. Previously only the bulk and the edge\r\nuniversality were established in this generality [arXiv:1804.07744], while cusp\r\nuniversality was proven only for Wigner-type matrices with independent entries\r\n[arXiv:1809.03971, arXiv:1811.04055]. As our main technical input, we prove an\r\noptimal local law at the cusp using the Zigzag strategy, a recursive tandem of\r\nthe characteristic flow method and a Green function comparison argument.\r\nMoreover, our proof of the optimal local law holds uniformly in the spectrum,\r\nthus also re-establishing universality of the local eigenvalue statistics in\r\nthe previously studied bulk [arXiv:1705.10661] and edge [arXiv:1804.07744]\r\nregimes."}],"publication":"arXiv","year":"2024","language":[{"iso":"eng"}],"external_id":{"arxiv":["2410.06813"]},"related_material":{"record":[{"relation":"later_version","id":"20322","status":"public"},{"status":"public","relation":"dissertation_contains","id":"20575"},{"relation":"dissertation_contains","id":"19540","status":"public"}]},"type":"preprint","doi":"10.48550/arXiv.2410.06813","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2410.06813","open_access":"1"}],"project":[{"_id":"62796744-2b32-11ec-9570-940b20777f1d","name":"Random matrices beyond Wigner-Dyson-Mehta","call_identifier":"H2020","grant_number":"101020331"}],"date_created":"2025-04-11T08:48:21Z","date_updated":"2026-04-07T12:37:11Z","_id":"19547","oa_version":"Preprint","month":"11","department":[{"_id":"LaEr"}],"arxiv":1,"ec_funded":1,"corr_author":"1","status":"public","user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","citation":{"ista":"Erdös L, Henheik SJ, Riabov V. Cusp universality for correlated random matrices. arXiv, 10.48550/arXiv.2410.06813.","mla":"Erdös, László, et al. “Cusp Universality for Correlated Random Matrices.” ArXiv, doi:10.48550/arXiv.2410.06813.","short":"L. Erdös, S.J. Henheik, V. Riabov, ArXiv (n.d.).","chicago":"Erdös, László, Sven Joscha Henheik, and Volodymyr Riabov. “Cusp Universality for Correlated Random Matrices.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2410.06813.","ama":"Erdös L, Henheik SJ, Riabov V. Cusp universality for correlated random matrices. arXiv. doi:10.48550/arXiv.2410.06813","apa":"Erdös, L., Henheik, S. J., & Riabov, V. (n.d.). Cusp universality for correlated random matrices. arXiv. https://doi.org/10.48550/arXiv.2410.06813","ieee":"L. Erdös, S. J. Henheik, and V. Riabov, “Cusp universality for correlated random matrices,” arXiv. ."},"date_published":"2024-11-03T00:00:00Z"},{"publication":"arXiv","department":[{"_id":"LaEr"}],"year":"2024","arxiv":1,"corr_author":"1","status":"public","external_id":{"arxiv":["2410.10809"]},"user_id":"8b945eb4-e2f2-11eb-945a-df72226e66a9","language":[{"iso":"eng"}],"citation":{"ista":"Henheik SJ, Wessel T. Response theory for locally gapped systems. arXiv, 10.48550/arXiv.2410.10809.","mla":"Henheik, Sven Joscha, and Tom Wessel. “Response Theory for Locally Gapped Systems.” ArXiv, doi:10.48550/arXiv.2410.10809.","short":"S.J. Henheik, T. Wessel, ArXiv (n.d.).","chicago":"Henheik, Sven Joscha, and Tom Wessel. “Response Theory for Locally Gapped Systems.” ArXiv, n.d. https://doi.org/10.48550/arXiv.2410.10809.","ama":"Henheik SJ, Wessel T. Response theory for locally gapped systems. arXiv. doi:10.48550/arXiv.2410.10809","apa":"Henheik, S. J., & Wessel, T. (n.d.). Response theory for locally gapped systems. arXiv. https://doi.org/10.48550/arXiv.2410.10809","ieee":"S. J. Henheik and T. Wessel, “Response theory for locally gapped systems,” arXiv. ."},"date_published":"2024-10-14T00:00:00Z","author":[{"last_name":"Henheik","id":"31d731d7-d235-11ea-ad11-b50331c8d7fb","orcid":"0000-0003-1106-327X","first_name":"Sven Joscha","full_name":"Henheik, Sven Joscha"},{"last_name":"Wessel","full_name":"Wessel, Tom","first_name":"Tom"}],"day":"14","related_material":{"record":[{"relation":"dissertation_contains","id":"19540","status":"public"}]},"type":"preprint","article_processing_charge":"No","tmp":{"image":"/images/cc_by.png","short":"CC BY (4.0)","name":"Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)","legal_code_url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"OA_place":"repository","oa":1,"doi":"10.48550/arXiv.2410.10809","publication_status":"draft","main_file_link":[{"url":"https://doi.org/10.48550/arXiv.2410.10809","open_access":"1"}],"title":"Response theory for locally gapped systems","oa_version":"Preprint","_id":"19551","month":"10","abstract":[{"lang":"eng","text":"We introduce a notion of a \\emph{local gap} for interacting many-body quantum lattice systems and prove the validity of response theory and Kubo's formula for localized perturbations in such settings.\r\nOn a high level, our result shows that the usual spectral gap condition, concerning the system as a whole, is not a necessary condition for understanding local properties of the system.\r\nMore precisely, we say that an equilibrium state ρ0 of a Hamiltonian H0 is locally gapped in Λgap⊂Λ, whenever the Liouvillian −i[H0,⋅] is almost invertible on local observables supported in Λgap when tested in ρ0.\r\nTo put this into context, we provide other alternative notions of a local gap and discuss their relations.\r\nThe validity of response theory is based on the construction of \\emph{non-equilibrium almost stationary states} (NEASSs).\r\nBy controlling locality properties of the NEASS construction, we show that response theory holds to any order, whenever the perturbation \\(\\epsilon V\\) acts in a region which is further than |logϵ| away from the non-gapped region Λ∖Λgap."}],"date_created":"2025-04-11T11:54:56Z","date_updated":"2026-04-07T12:37:11Z"}]