{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T20:22:42Z","timestamp":1740169362621,"version":"3.37.3"},"reference-count":68,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"National Science Center, Poland","award":["2022\/45\/B\/ST6\/02817"],"award-info":[{"award-number":["2022\/45\/B\/ST6\/02817"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2024]]},"DOI":"10.1109\/access.2024.3406261","type":"journal-article","created":{"date-parts":[[2024,5,27]],"date-time":"2024-05-27T17:22:54Z","timestamp":1716830574000},"page":"76367-76378","source":"Crossref","is-referenced-by-count":0,"title":["Toward Unsupervised Visual Reasoning: Do Off-the-Shelf Features Know How to Reason?"],"prefix":"10.1109","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7785-2277","authenticated-orcid":false,"given":"Monika","family":"Wysocza\u0144ska","sequence":"first","affiliation":[{"name":"Faculty of Electronics and Information Technology, Warsaw University of Technology, Warsaw, Poland"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tom","family":"Monnier","sequence":"additional","affiliation":[{"name":"LIGM, École des Ponts, CNRS, Univ Gustave Eiffel, Marne-la-Vallée, France"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1486-8906","authenticated-orcid":false,"given":"Tomasz","family":"Trzci\u0144ski","sequence":"additional","affiliation":[{"name":"Faculty of Electronics and Information Technology, Warsaw University of Technology, Warsaw, Poland"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6296-4222","authenticated-orcid":false,"given":"David","family":"Picard","sequence":"additional","affiliation":[{"name":"LIGM, École des Ponts, CNRS, Univ Gustave Eiffel, Marne-la-Vallée, France"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00951"},{"key":"ref2","article-title":"DINOv2: Learning robust visual features without supervision","author":"Oquab","year":"2024","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00950"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00393"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.5555\/3524938.3525087"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.5555\/3495724.3497510"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00975"},{"article-title":"Visual representation learning does not generalize strongly within the same domain","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Schott","key":"ref8"},{"key":"ref9","article-title":"MONet: Unsupervised scene decomposition and representation","author":"Burgess","year":"2019","journal-title":"arXiv:1901.11390"},{"key":"ref10","first-page":"2424","article-title":"Multi-object representation learning with iterative variational inference","volume-title":"Proc. ICML","volume":"97","author":"Greff"},{"key":"ref11","first-page":"11525","article-title":"Object-centric learning with slot attention","volume-title":"Proc. Conf. Neural Inf. Process. Syst. (NeurIPS)","volume":"33","author":"Locatello"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00852"},{"key":"ref13","article-title":"Clevrtex: A texture-rich benchmark for unsupervised multi-object segmentation","volume-title":"Proc. 35th Conf. Neural Inf. Process. Syst. Datasets Benchmarks Track (Round)","volume":"1","author":"Karazija"},{"article-title":"Measuring CLEVRness: Black-box testing of visual reasoning models","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Mouselinos","key":"ref14"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.5220\/0011895400003417"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA40945.2020.9197332"},{"article-title":"PaLI: A jointly-scaled multilingual language-image model","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Chen","key":"ref17"},{"key":"ref18","first-page":"23716","article-title":"Flamingo: A visual language model for few-shot learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Alayrac"},{"key":"ref19","first-page":"19730","article-title":"BLIP-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Li"},{"key":"ref20","article-title":"Visual instruction tuning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"36","author":"Liu"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref22","article-title":"ABC-CNN: An attention based convolutional neural network for visual question answering","author":"Chen","year":"2015","journal-title":"arXiv:1511.05960"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11671"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00836"},{"key":"ref25","article-title":"Compositional attention networks for machine reasoning","author":"Hudson","year":"2018","journal-title":"arXiv:1803.03067"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00636"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00189"},{"article-title":"The neuro-symbolic concept learner: Interpreting scenes, words, and sentences from natural supervision","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Mao","key":"ref28"},{"article-title":"Disentangling 3D prototypical networks for few-shot concept learning","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Prabhudesai","key":"ref29"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01438"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i3.20215"},{"article-title":"Neural-symbolic VQA: Disentangling reasoning from vision and language understanding","volume-title":"Proc. NeurIPS","author":"Yi","key":"ref32"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00519"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2021.10.016"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","volume-title":"Proc. ICLR","author":"Dosovitskiy","key":"ref36"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58577-8_7"},{"key":"ref38","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. Int. Conf. Mach. Learn.","volume":"139","author":"Radford"},{"article-title":"ViLBERT: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks","volume-title":"Proc. NeurIPS","author":"Lu","key":"ref39"},{"article-title":"Deep learning for symbolic mathematics","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Lample","key":"ref40"},{"article-title":"Teaching temporal logics to neural networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Hahn","key":"ref41"},{"issue":"75","key":"ref42","first-page":"1","article-title":"Attention is turing-complete","volume":"22","author":"P\u00e9rez","year":"2021","journal-title":"J. Mach. Learn. Res."},{"key":"ref43","article-title":"Attention over learned object embeddings enables complex visual reasoning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Ding"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00180"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW.2014.131"},{"key":"ref46","first-page":"4904","article-title":"Scaling up visual and vision-language representation learning with noisy text supervision","volume-title":"Proc. 38th Int. Conf. Mach. Learn.","volume":"139","author":"Jia"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i21.30339"},{"key":"ref48","article-title":"Clipcap: Clip prefix for image captioning","author":"Mokady","year":"2021","journal-title":"arXiv:2111.09734"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.02080"},{"key":"ref50","article-title":"CLIP-DINOiser: Teaching CLIP a few DINO tricks for open-vocabulary semantic segmentation","author":"Wysocza\u0144ska","year":"2023","journal-title":"arXiv:2312.12359"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/WACV57701.2024.00143"},{"article-title":"Binding via reconstruction clustering","volume-title":"Proc. ICLR Workshops","author":"Greff","key":"ref52"},{"article-title":"Neural expectation maximization","volume-title":"Proc. NeurIPS","author":"Greff","key":"ref53"},{"article-title":"GENESIS: Generative scene inference and sampling with object-centric latent representations","volume-title":"Proc. ICLR","author":"Engelcke","key":"ref54"},{"article-title":"Spatial transformer networks","volume-title":"Proc. NeurIPS","author":"Jaderberg","key":"ref55"},{"article-title":"Attend, infer, repeat: Fast scene understanding with generative models","volume-title":"Proc. NeurIPS","author":"Eslami","key":"ref56"},{"article-title":"Sequential attend, infer, repeat: Generative modelling of moving objects","volume-title":"Proc. NeurIPS","author":"Kosiorek","key":"ref57"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33013412"},{"article-title":"SPACE: Unsupervised object-oriented scene representation via spatial attention and decomposition","volume-title":"Proc. ICLR","author":"Lin","key":"ref59"},{"article-title":"MarioNette: Self-supervised sprite learning","volume-title":"Proc. NeurIPS","author":"Smirnov","key":"ref60"},{"key":"ref61","article-title":"RoBERTa: A robustly optimized BERT pretraining approach","author":"Liu","year":"1907","journal-title":"arXiv:1907.11692"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.215"},{"key":"ref63","article-title":"CLEVR-Math: A dataset for compositional language, visual and mathematical reasoning","author":"Lindstr\u00f6m","year":"2022","journal-title":"arXiv:2208.05358"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref65","article-title":"HuggingFace\u2019s transformers: State-of-the-art natural language processing","author":"Wolf","year":"2019","journal-title":"arXiv:1910.03771"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.5555\/1953048.2078195"},{"key":"ref67","first-page":"8024","article-title":"PyTorch: An imperative style, high-performance deep learning library","volume-title":"Proc. NIPS","author":"Paszke"},{"key":"ref68","article-title":"Decoupled weight decay regularization","author":"Loshchilov","year":"2017","journal-title":"arXiv:1711.05101"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6287639\/10380310\/10539985.pdf?arnumber=10539985","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,6,5]],"date-time":"2024-06-05T17:57:54Z","timestamp":1717610274000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10539985\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024]]},"references-count":68,"URL":"https:\/\/doi.org\/10.1109\/access.2024.3406261","relation":{},"ISSN":["2169-3536"],"issn-type":[{"type":"electronic","value":"2169-3536"}],"subject":[],"published":{"date-parts":[[2024]]}}}