[{"date_created":"2023-08-22T14:22:04Z","year":"2023","article_processing_charge":"No","oa":1,"external_id":{"arxiv":["2210.01738"]},"acknowledgement":"AN, MF, and FL partially worked on ASIF when they were at Amazon Web Services in Tübingen,\r\nGermany. This paper is financially supported by the PRIN 2020 project no.2020TA3K9N (LEGO.AI), PNRR MUR project PE0000013-FAIR, and ERC Grant no.802554 (SPECGEO).","citation":{"apa":"Norelli, A., Fumero, M., Maiorca, V., Moschella, L., Rodolà, E., &#38; Locatello, F. (2023). ASIF: Coupled data turns unimodal models to multimodal without training. In <i>37th Conference on Neural Information Processing Systems</i> (Vol. 36, pp. 15303–15319). New Orleans, LA, United States: Neural Information Processing Systems Foundation.","ieee":"A. Norelli, M. Fumero, V. Maiorca, L. Moschella, E. Rodolà, and F. Locatello, “ASIF: Coupled data turns unimodal models to multimodal without training,” in <i>37th Conference on Neural Information Processing Systems</i>, New Orleans, LA, United States, 2023, vol. 36, pp. 15303–15319.","ista":"Norelli A, Fumero M, Maiorca V, Moschella L, Rodolà E, Locatello F. 2023. ASIF: Coupled data turns unimodal models to multimodal without training. 37th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 36, 15303–15319.","chicago":"Norelli, Antonio, Marco Fumero, Valentino Maiorca, Luca Moschella, Emanuele Rodolà, and Francesco Locatello. “ASIF: Coupled Data Turns Unimodal Models to Multimodal without Training.” In <i>37th Conference on Neural Information Processing Systems</i>, 36:15303–19. Neural Information Processing Systems Foundation, 2023.","mla":"Norelli, Antonio, et al. “ASIF: Coupled Data Turns Unimodal Models to Multimodal without Training.” <i>37th Conference on Neural Information Processing Systems</i>, vol. 36, Neural Information Processing Systems Foundation, 2023, pp. 15303–19.","short":"A. Norelli, M. Fumero, V. Maiorca, L. Moschella, E. Rodolà, F. Locatello, in:, 37th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2023, pp. 15303–15319.","ama":"Norelli A, Fumero M, Maiorca V, Moschella L, Rodolà E, Locatello F. ASIF: Coupled data turns unimodal models to multimodal without training. In: <i>37th Conference on Neural Information Processing Systems</i>. Vol 36. Neural Information Processing Systems Foundation; 2023:15303-15319."},"intvolume":"        36","volume":36,"day":"04","alternative_title":["Advances in Neural Information Processing Systems"],"arxiv":1,"author":[{"full_name":"Norelli, Antonio","last_name":"Norelli","first_name":"Antonio"},{"last_name":"Fumero","first_name":"Marco","full_name":"Fumero, Marco"},{"first_name":"Valentino","last_name":"Maiorca","full_name":"Maiorca, Valentino"},{"first_name":"Luca","last_name":"Moschella","full_name":"Moschella, Luca"},{"full_name":"Rodolà, Emanuele","last_name":"Rodolà","first_name":"Emanuele"},{"first_name":"Francesco","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","last_name":"Locatello","orcid":"0000-0002-4850-0683","full_name":"Locatello, Francesco"}],"abstract":[{"lang":"eng","text":"CLIP proved that aligning visual and language spaces is key to solving many vision tasks without explicit training, but required to train image and text encoders from scratch on a huge dataset. LiT improved this by only training the text encoder and using a pre-trained vision network. In this paper, we show that a common space can be created without any training at all, using single-domain encoders (trained with or without supervision) and a much smaller amount of image-text pairs. Furthermore, our model has unique properties. Most notably, deploying a new version with updated training samples can be done in a matter of seconds. Additionally, the representations in the common space are easily interpretable as every dimension corresponds to the similarity of the input to a unique entry in the multimodal dataset. Experiments on standard zero-shot visual benchmarks demonstrate the typical transfer ability of image-text models. Overall, our method represents a simple yet surprisingly strong baseline for foundation multi-modal models, raising important questions on their data efficiency and on the role of retrieval in machine learning."}],"date_updated":"2025-05-14T11:28:52Z","title":"ASIF: Coupled data turns unimodal models to multimodal without training","department":[{"_id":"FrLo"}],"OA_type":"green","file":[{"creator":"dernst","date_updated":"2025-02-04T12:16:13Z","access_level":"open_access","file_id":"18994","checksum":"e51c90300b92d7135050da5c9e3a8015","file_size":12648978,"file_name":"2023_NeurIPS_Fumero.pdf","success":1,"relation":"main_file","date_created":"2025-02-04T12:16:13Z","content_type":"application/pdf"}],"publication_identifier":{"isbn":["9781713899921"]},"corr_author":"1","quality_controlled":"1","file_date_updated":"2025-02-04T12:16:13Z","status":"public","has_accepted_license":"1","date_published":"2023-10-04T00:00:00Z","page":"15303-15319","oa_version":"Preprint","_id":"14216","ddc":["000"],"month":"10","publisher":"Neural Information Processing Systems Foundation","type":"conference","conference":{"location":"New Orleans, LA, United States","name":"NeurIPS: Neural Information Processing Systems","start_date":"2023-12-12","end_date":"2023-12-14"},"related_material":{"link":[{"url":"https://github.com/noranta4/ASIF","relation":"software"}]},"publication":"37th Conference on Neural Information Processing Systems","language":[{"iso":"eng"}],"publication_status":"published","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87"},{"alternative_title":["Advances in Neural Information Processing Systems"],"day":"27","volume":36,"citation":{"ama":"Zhu Z, Locatello F, Cevher V. Sample complexity bounds for score-matching: Causal discovery and generative modeling. In: <i>37th Conference on Neural Information Processing Systems</i>. Vol 36. Neural Information Processing Systems Foundation; 2023:3325-3337. doi:<a href=\"https://doi.org/10.52202/075280-0147\">10.52202/075280-0147</a>","ista":"Zhu Z, Locatello F, Cevher V. 2023. Sample complexity bounds for score-matching: Causal discovery and generative modeling. 37th Conference on Neural Information Processing Systems. NeurIPS: Neural Information Processing Systems, Advances in Neural Information Processing Systems, vol. 36, 3325–3337.","apa":"Zhu, Z., Locatello, F., &#38; Cevher, V. (2023). Sample complexity bounds for score-matching: Causal discovery and generative modeling. In <i>37th Conference on Neural Information Processing Systems</i> (Vol. 36, pp. 3325–3337). New Orleans, LA, United States: Neural Information Processing Systems Foundation. <a href=\"https://doi.org/10.52202/075280-0147\">https://doi.org/10.52202/075280-0147</a>","ieee":"Z. Zhu, F. Locatello, and V. Cevher, “Sample complexity bounds for score-matching: Causal discovery and generative modeling,” in <i>37th Conference on Neural Information Processing Systems</i>, New Orleans, LA, United States, 2023, vol. 36, pp. 3325–3337.","mla":"Zhu, Zhenyu, et al. “Sample Complexity Bounds for Score-Matching: Causal Discovery and Generative Modeling.” <i>37th Conference on Neural Information Processing Systems</i>, vol. 36, Neural Information Processing Systems Foundation, 2023, pp. 3325–37, doi:<a href=\"https://doi.org/10.52202/075280-0147\">10.52202/075280-0147</a>.","short":"Z. Zhu, F. Locatello, V. Cevher, in:, 37th Conference on Neural Information Processing Systems, Neural Information Processing Systems Foundation, 2023, pp. 3325–3337.","chicago":"Zhu, Zhenyu, Francesco Locatello, and Volkan Cevher. “Sample Complexity Bounds for Score-Matching: Causal Discovery and Generative Modeling.” In <i>37th Conference on Neural Information Processing Systems</i>, 36:3325–37. Neural Information Processing Systems Foundation, 2023. <a href=\"https://doi.org/10.52202/075280-0147\">https://doi.org/10.52202/075280-0147</a>."},"intvolume":"        36","arxiv":1,"researchdata_availability":"no","abstract":[{"text":"This paper provides statistical sample complexity bounds for score-matching and\r\nits applications in causal discovery. We demonstrate that accurate estimation of the\r\nscore function is achievable by training a standard deep ReLU neural network using\r\nstochastic gradient descent. We establish bounds on the error rate of recovering\r\ncausal relationships using the score-matching-based causal discovery method of\r\nRolland et al. [2022], assuming a sufficiently good estimation of the score function.\r\nFinally, we analyze the upper bound of score-matching estimation within the scorebased generative modeling, which has been applied for causal discovery but is also\r\nof independent interest within the domain of generative models.y","lang":"eng"}],"author":[{"first_name":"Zhenyu","last_name":"Zhu","full_name":"Zhu, Zhenyu"},{"full_name":"Locatello, Francesco","orcid":"0000-0002-4850-0683","last_name":"Locatello","id":"26cfd52f-2483-11ee-8040-88983bcc06d4","first_name":"Francesco"},{"full_name":"Cevher, Volkan","first_name":"Volkan","last_name":"Cevher"}],"title":"Sample complexity bounds for score-matching: Causal discovery and generative modeling","date_updated":"2026-09-17T12:59:35Z","article_processing_charge":"No","year":"2023","date_created":"2024-02-07T15:11:11Z","supplementarymaterial":"yes","acknowledgement":"We are thankful to the reviewers for providing constructive feedback and Kun Zhang and Dominik\r\nJanzing for helpful discussion on the special case of deterministic children. This work was supported\r\nby Hasler Foundation Program: Hasler Responsible AI (project number 21043). This work was\r\nsupported by the Swiss National Science Foundation (SNSF) under grant number 200021_205011.\r\nFrancesco Locatello did not contribute to this work at Amazon. ","external_id":{"arxiv":["2310.18123"]},"oa":1,"page":"3325-3337","ddc":["000"],"month":"10","oa_version":"Published Version","_id":"14953","scopus_import":"1","conference":{"location":"New Orleans, LA, United States","name":"NeurIPS: Neural Information Processing Systems","start_date":"2023-12-12","end_date":"2023-12-16"},"type":"conference","publisher":"Neural Information Processing Systems Foundation","publication_status":"published","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","publication":"37th Conference on Neural Information Processing Systems","language":[{"iso":"eng"}],"fulldoi":"https://doi.org/10.52202/075280-0147","OA_type":"gold","department":[{"_id":"FrLo"}],"OA_place":"publisher","quality_controlled":"1","corr_author":"1","file":[{"relation":"main_file","success":1,"content_type":"application/pdf","date_created":"2026-08-13T07:02:51Z","file_id":"22700","access_level":"open_access","creator":"dernst","date_updated":"2026-08-13T07:02:51Z","file_name":"2023_Neurips_Zhu.pdf","checksum":"6f10adadefceb3bb50ad0b2637493381","file_size":305362}],"doi":"10.52202/075280-0147","publication_identifier":{"eissn":["1049-5258"],"isbn":["9781713899921"]},"file_date_updated":"2026-08-13T07:02:51Z","status":"public","date_published":"2023-10-27T00:00:00Z","das_tickbox":"0","has_accepted_license":"1"}]