{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,23]],"date-time":"2026-06-23T08:09:43Z","timestamp":1782202183022,"version":"3.54.5"},"reference-count":323,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2025]]},"DOI":"10.1109\/access.2025.3589514","type":"journal-article","created":{"date-parts":[[2025,7,15]],"date-time":"2025-07-15T17:43:25Z","timestamp":1752601405000},"page":"125537-125569","source":"Crossref","is-referenced-by-count":1,"title":["Scale Is Not All You Need: Revisiting the Biomimetic Roots of Deep Learning to Overcome Fundamental Limitations"],"prefix":"10.1109","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2247-0770","authenticated-orcid":false,"given":"Malek","family":"Wahidi","sequence":"first","affiliation":[{"name":"Department of Computer Engineering, University of Balamand, Koura, Lebanon"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9401-6972","authenticated-orcid":false,"given":"Anthony","family":"Rizk","sequence":"additional","affiliation":[{"name":"Innovation Laboratory, IDEALworks GmbH, Munich, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4627-9778","authenticated-orcid":false,"given":"Rodrigue","family":"Imad","sequence":"additional","affiliation":[{"name":"Department of Computer Engineering, University of Balamand, Koura, Lebanon"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/bf02478259"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1037\/h0042519"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1113\/jphysiol.1962.sp006837"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/0893-6080(88)90014-7"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/tssc.1969.300225"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"ref7","volume-title":"Reinforcement Learning, Second Edition: An Introduction","author":"Sutton","year":"2018"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-023-37180-x"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1038\/323533a0"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1017\/S0140525X16001837"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2019.08.034"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2017.06.011"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1098\/rspa.2021.0068"},{"key":"ref14","volume-title":"The Scaling Hypothesis","year":"2025"},{"key":"ref15","article-title":"Deep learning: A critical appraisal","author":"Marcus","year":"2018","journal-title":"arXiv:1801.00631"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1016\/j.cub.2014.08.026"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.21428\/bf6fb269.1f033948"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1016\/j.tics.2023.10.002"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.3389\/fncom.2023.1092185"},{"key":"ref20","article-title":"Spiking neural networks and bio-inspired supervised deep learning: A survey","author":"Lagani","year":"2023","journal-title":"arXiv:2307.16235"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2022\/774"},{"key":"ref22","article-title":"A survey on brain-inspired deep learning via predictive coding","author":"Salvatori","year":"2025","journal-title":"arXiv:2308.07870"},{"key":"ref23","volume-title":"AI and Compute","year":"2024"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1038\/nature24270"},{"key":"ref25","volume-title":"Training Compute of Frontier AI Models Grows by 4\u20135x Per Year","author":"Sevilla","year":"2024"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ijcnn55064.2022.9891914"},{"key":"ref27","volume-title":"AI Index Report 2024\u2014Artificial Intelligence Index","year":"2024"},{"key":"ref28","article-title":"Trends in AI supercomputers","author":"Pilz","year":"2025","journal-title":"arXiv:2504.16026"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1038\/530144a"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/mcse.2017.29"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/mspec.2021.9563954"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1355"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/s13369-024-09920-8"},{"key":"ref34","article-title":"On the opportunities and risks of foundation models","author":"Bommasani","year":"2021","journal-title":"arXiv:2108.07258"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.97"},{"key":"ref36","article-title":"DINOv2: Learning robust visual features without supervision","author":"Oquab","year":"2023","journal-title":"arXiv:2304.07193"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"ref38","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. 38th Int. Conf. Mach. Learn.","author":"Radford"},{"key":"ref39","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Brown"},{"key":"ref40","volume-title":"Will We Run Out of Data? Limits of LLM Scaling Based on Human-Generated Data","author":"Villalobos","year":"2024"},{"key":"ref41","volume-title":"Introducing Meta Llama 3: The Most Capable Openly Available LLM to Date","year":"2024"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-47560-3"},{"key":"ref43","article-title":"Best practices and lessons learned on synthetic data","author":"Liu","year":"2024","journal-title":"arXiv:2404.07503"},{"key":"ref44","article-title":"Machine learning for synthetic data generation: A review","author":"Lu","year":"2023","journal-title":"arXiv:2302.04062"},{"key":"ref45","article-title":"The llama 3 herd of models","author":"Grattafiori","year":"2024","journal-title":"arXiv:2407.21783"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-03051-4"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06747-5"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-019-1724-z"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1145\/3381831"},{"key":"ref50","article-title":"Highlighting the importance of reducing research bias and carbon emissions in CNNs","author":"Badar","year":"2021","journal-title":"arXiv:2106.03242"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.3389\/fnana.2024.1374864"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1201895109"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1016\/s0960-9822(03)00135-0"},{"key":"ref54","volume-title":"Human Learning in Atari","author":"Tsividis","year":"2017"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2107022118"},{"key":"ref56","article-title":"No \u2019Zero-shot\u2019 without exponential data: Pretraining concept frequency determines multimodal model performance","author":"Udandarao","year":"2024","journal-title":"arXiv:2404.04125"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i16.29808"},{"key":"ref58","article-title":"Impact of pretraining term frequencies on few-shot reasoning","author":"Razeghi","year":"2022","journal-title":"arXiv:2202.07206"},{"key":"ref59","article-title":"A survey on evaluation of out-of-distribution generalization","author":"Yu","year":"2024","journal-title":"arXiv:2403.01874"},{"key":"ref60","article-title":"Towards out-of-distribution generalization: A survey","author":"Liu","year":"2021","journal-title":"arXiv:2108.13624"},{"key":"ref61","first-page":"18583","article-title":"Measuring robustness to natural distribution shifts in image classification","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Taori"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1145\/3571724"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1145\/3711118"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-019-11786-6"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1038\/nature16961"},{"key":"ref66","article-title":"Intriguing properties of neural networks","author":"Szegedy","year":"2013","journal-title":"arXiv:1312.6199"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1038\/d41586-019-03013-5"},{"key":"ref68","article-title":"Explaining and harnessing adversarial examples","author":"Goodfellow","year":"2014","journal-title":"arXiv:1412.6572"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1007\/s13748-021-00269-9"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/iccvw.2019.00134"},{"key":"ref71","article-title":"ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness","author":"Geirhos","year":"2018","journal-title":"arXiv:1811.12231"},{"key":"ref72","first-page":"19000","article-title":"The origins and prevalence of texture bias in convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Hermann"},{"key":"ref73","first-page":"23296","article-title":"Intriguing properties of vision transformers","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Naseer"},{"key":"ref74","article-title":"A systematic review of robustness in deep learning for computer Vision: Mind the gap?","author":"Drenkow","year":"2022","journal-title":"arXiv:2112.00639"},{"key":"ref75","first-page":"1","article-title":"Generalisation in humans and deep neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Geirhos"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1145\/3501714.3501755"},{"key":"ref77","first-page":"1","article-title":"Adversarial examples are not bugs, they are features","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ilyas"},{"key":"ref78","article-title":"Robustness may be at odds with accuracy","author":"Tsipras","year":"2018","journal-title":"arXiv:1805.12152"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/ICCCN.2017.8038465"},{"key":"ref80","first-page":"1","article-title":"Attention is all you need","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Vaswani"},{"key":"ref81","article-title":"Can in-context learning really generalize to out-of-distribution tasks?","author":"Wang","year":"2024","journal-title":"arXiv:2410.09695"},{"key":"ref82","article-title":"A closer look at in-context learning under distribution shifts","volume-title":"Workshop Efficient Syst. Found. Models (ICML)","author":"Ahuja"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/tkde.2018.2876857"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2021.3117837"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"ref86","article-title":"Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity","author":"Fedus","year":"2021","journal-title":"arXiv:2101.03961"},{"key":"ref87","article-title":"DeepSeek-V3 technical report","volume-title":"arXiv:2412.19437","author":"Liu","year":"2024"},{"key":"ref88","article-title":"Kimi-VL technical report","volume-title":"arXiv:2504.07491","author":"Team","year":"2025"},{"key":"ref89","article-title":"Mixtral of experts","author":"Jiang","year":"2024","journal-title":"arXiv:2401.04088"},{"key":"ref90","article-title":"Mixture-of-depths: Dynamically allocating compute in transformer-based language models","author":"Raposo","year":"2024","journal-title":"arXiv:2404.02258"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.681"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.1109\/icpr.2016.7900006"},{"key":"ref93","first-page":"17456","article-title":"Confident adaptive language modeling","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Schuster"},{"key":"ref94","article-title":"DeepSeek-r1: Incentivizing reasoning capability in LLMs via reinforcement learning","author":"Guo","year":"2025","journal-title":"arXiv:2501.12948"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2013.10.013"},{"key":"ref96","doi-asserted-by":"publisher","DOI":"10.1016\/s0925-2312(99)00063-6"},{"key":"ref97","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-63755-5"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1008215"},{"key":"ref99","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2007.05.019"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.1111\/nyas.14320"},{"key":"ref101","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-022-00452-0"},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1800755115"},{"key":"ref103","doi-asserted-by":"publisher","DOI":"10.1016\/s1364-6613(99)01294-2"},{"key":"ref104","doi-asserted-by":"publisher","DOI":"10.1016\/s0079-7421(08)60536-8"},{"key":"ref105","volume-title":"Unsupervised Feature Learning and Deep Learning: A Review and New Perspectives","author":"Bengio","year":"2012"},{"key":"ref106","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-019-05855-6"},{"key":"ref107","doi-asserted-by":"publisher","DOI":"10.1109\/msp.2021.3134634"},{"key":"ref108","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-021-05961-4"},{"key":"ref109","article-title":"Reinforcement learning in practice: Opportunities and challenges","author":"Li","year":"2022","journal-title":"arXiv:2202.11296"},{"key":"ref110","article-title":"Gemini: A family of highly capable multimodal models","author":"Team","year":"2023","journal-title":"arXiv:2312.11805"},{"key":"ref111","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"arXiv:1810.04805"},{"key":"ref112","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv:2307.09288"},{"key":"ref113","article-title":"Where are we in the search for an artificial visual cortex for embodied intelligence?","author":"Majumdar","journal-title":"arXiv:2303.18240"},{"key":"ref114","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2023.xix.025"},{"key":"ref115","article-title":"RT-2: Vision-language-action models transfer web knowledge to robotic control","author":"Brohan","year":"2023","journal-title":"arXiv:2307.15818"},{"key":"ref116","doi-asserted-by":"publisher","DOI":"10.1109\/jbhi.2024.3350666"},{"key":"ref117","doi-asserted-by":"publisher","DOI":"10.1109\/tcbb.2023.3254163"},{"key":"ref118","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52729.2023.01499"},{"key":"ref119","volume-title":"V-JEPA: Latent Video Prediction for Visual Representation Learning","author":"Bardes","year":"2023"},{"key":"ref120","doi-asserted-by":"publisher","DOI":"10.1038\/s44159-023-00166-z"},{"key":"ref121","volume-title":"Improving Language Understanding By Generative Pre-Training","author":"Radford","year":"2018"},{"key":"ref122","volume-title":"Language Models are Unsupervised Multitask Learners","author":"Radford","year":"2019"},{"key":"ref123","doi-asserted-by":"publisher","DOI":"10.1162\/neco_a_00999"},{"key":"ref124","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2021.05.008"},{"key":"ref125","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-023-10562-9"},{"key":"ref126","doi-asserted-by":"publisher","DOI":"10.1109\/tnn.1998.712192"},{"key":"ref127","doi-asserted-by":"publisher","DOI":"10.1177\/17298814211007305"},{"key":"ref128","doi-asserted-by":"publisher","DOI":"10.1177\/0278364920987859"},{"key":"ref129","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2022.03.003"},{"key":"ref130","doi-asserted-by":"publisher","DOI":"10.1080\/14697688.2019.1655928"},{"key":"ref131","doi-asserted-by":"publisher","DOI":"10.1145\/3159652.3176182"},{"key":"ref132","doi-asserted-by":"publisher","DOI":"10.3389\/fnsys.2021.772641"},{"key":"ref133","doi-asserted-by":"publisher","DOI":"10.1109\/jproc.2021.3058954"},{"key":"ref134","doi-asserted-by":"publisher","DOI":"10.1007\/s11633-022-1362-z"},{"key":"ref135","article-title":"Spurious correlations in machine learning: A survey","author":"Ye","year":"2024","journal-title":"arXiv:2402.12715"},{"key":"ref136","volume-title":"Understanding Rare Spurious Correlations in Neural Networks","author":"Yang","year":"2022"},{"key":"ref137","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-023-01916-5"},{"key":"ref138","article-title":"Measuring the tendency of CNNs to learn surface statistical regularities","author":"Jo","year":"2017","journal-title":"arXiv:1711.11561"},{"key":"ref139","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2023\/375"},{"key":"ref140","article-title":"Changing answer order can decrease MMLU accuracy","author":"Gupta","year":"2024","journal-title":"arXiv:2406.19470"},{"key":"ref141","first-page":"4555","article-title":"Selecting data augmentation for simulating interventions","volume-title":"Proc. 38th Int. Conf. Mach. Learn.","author":"Ilse"},{"key":"ref142","first-page":"8346","article-title":"An investigation of why overparameterization exacerbates spurious correlations","volume-title":"Proc. 37th Int. Conf. Mach. Learn.","author":"Sagawa"},{"key":"ref143","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1459"},{"key":"ref144","volume-title":"On the Paradox of Generalizable Logical Reasoning in Large Language Models","author":"Tang","year":"2023"},{"key":"ref145","first-page":"1819","article-title":"Reasoning or reciting? Exploring the capabilities and limitations of language models through counterfactual tasks","volume-title":"Proc. Conf. North Amer. Chapter Assoc. Comput. Linguistics, Hum. Lang. Technol.","author":"Wu"},{"key":"ref146","article-title":"GSM-symbolic: Understanding the limitations of mathematical reasoning in large language models","author":"Mirzadeh","year":"2024","journal-title":"arXiv:2410.05229"},{"key":"ref147","volume-title":"Causal Parrots: Large Language Models May Talk Causality But Are Not Causal","author":"Ze\u010devi\u0107","year":"2023"},{"key":"ref148","article-title":"The reversal curse: LLMs trained on \u2019A is B\u2019 fail to learn \u2019B is A","volume-title":"12th Int. Conf. Learn. Represent","author":"Berglund"},{"key":"ref149","article-title":"MATH-perturb: Benchmarking LLMs\u2019 math reasoning abilities against hard perturbations","author":"Huang","year":"2025","journal-title":"arXiv:2502.06453"},{"key":"ref150","article-title":"Recitation over reasoning: How cutting-edge language models can fail on elementary school-level reasoning problems?","author":"Yan","year":"2025","journal-title":"arXiv:2504.00509"},{"key":"ref151","article-title":"Hallucination is inevitable: An innate limitation of large language models","author":"Xu","year":"2024","journal-title":"arXiv:2401.11817"},{"key":"ref152","article-title":"LLMs will always hallucinate, and we need to live with this","author":"Banerjee","year":"2024","journal-title":"arXiv:2409.05746"},{"key":"ref153","article-title":"Evaluating the world model implicit in a generative model","author":"Vafa","year":"2024","journal-title":"arXiv:2406.03689"},{"key":"ref154","article-title":"How far is video generation from world model: A physical law perspective","author":"Kang","year":"2024","journal-title":"arXiv:2411.02385"},{"key":"ref155","doi-asserted-by":"publisher","DOI":"10.1111\/j.1468-5884.2009.00396.x"},{"key":"ref156","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-8721.2006.00458.x"},{"key":"ref157","doi-asserted-by":"publisher","DOI":"10.1093\/acprof:oso\/9780199216093.003.0020"},{"key":"ref158","doi-asserted-by":"publisher","DOI":"10.1016\/0893-6080(89)90020-8"},{"key":"ref159","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2021.07.002"},{"key":"ref160","doi-asserted-by":"publisher","DOI":"10.1038\/sj.embor.7400789"},{"key":"ref161","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroscience.2021.10.001"},{"key":"ref162","doi-asserted-by":"publisher","DOI":"10.1016\/j.conb.2010.07.009"},{"key":"ref163","volume-title":"Deep Learning","author":"Goodfellow","year":"2016"},{"key":"ref164","doi-asserted-by":"publisher","DOI":"10.1146\/annurev.neuro.28.061604.135703"},{"key":"ref165","article-title":"Can single neurons solve MNIST? The computational power of biological dendritic trees","author":"Jones","year":"2020","journal-title":"arXiv:2009.01269"},{"key":"ref166","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2021.09.044"},{"key":"ref167","doi-asserted-by":"publisher","DOI":"10.1038\/s41583-023-00756-z"},{"key":"ref168","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/14636.001.0001"},{"key":"ref169","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.abq7592"},{"key":"ref170","doi-asserted-by":"publisher","DOI":"10.1109\/icra.2019.8793840"},{"key":"ref171","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-020-00237-3"},{"key":"ref172","article-title":"Gaussian splatting to real world flight navigation transfer with liquid networks","author":"Quach","year":"2024","journal-title":"arXiv:2406.15149"},{"key":"ref173","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.adc8892"},{"key":"ref174","doi-asserted-by":"publisher","DOI":"10.1007\/s11063-023-11194-4"},{"key":"ref175","doi-asserted-by":"publisher","DOI":"10.1007\/s11063-024-11640-x"},{"key":"ref176","article-title":"Exploring liquid neural networks on Loihi-2","author":"Pawlak","year":"2024","journal-title":"arXiv:2407.20590"},{"key":"ref177","article-title":"LTC-SE: Expanding the potential of liquid time-constant neural networks for scalable AI and embedded systems","author":"Bidollahkhani","year":"2023","journal-title":"arXiv:2304.08691"},{"key":"ref178","article-title":"Recurrent convolution for compact and cost-adjustable neural networks: An empirical study","author":"Zhang","year":"2019","journal-title":"arXiv:1902.09809"},{"key":"ref179","doi-asserted-by":"publisher","DOI":"10.1109\/access.2023.3245808"},{"key":"ref180","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr.2015.7298958"},{"key":"ref181","article-title":"A review of neuroscience-inspired machine learning","author":"Ororbia","year":"2024","journal-title":"arXiv:2403.18929"},{"key":"ref182","volume-title":"Sample-efficient Image Segmentation Through Recurrence","author":"Linsley","year":"2018"},{"key":"ref183","doi-asserted-by":"publisher","DOI":"10.1126\/science.aag2612"},{"key":"ref184","doi-asserted-by":"publisher","DOI":"10.1038\/s41593-023-01514-1"},{"key":"ref185","doi-asserted-by":"publisher","DOI":"10.1109\/ijcnn48605.2020.9207382"},{"key":"ref186","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1007805"},{"key":"ref187","first-page":"12425","article-title":"Causal navigation by continuous-time neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Vorbach"},{"key":"ref188","doi-asserted-by":"publisher","DOI":"10.3850\/iahr-hic2483430201-274"},{"key":"ref189","doi-asserted-by":"publisher","DOI":"10.1109\/icieam60818.2024.10553901"},{"key":"ref190","first-page":"1","article-title":"Brain-inspired robust vision using convolutional neural networks with feedback","volume-title":"Proc. NeurIPS","author":"Huang"},{"key":"ref191","first-page":"535","article-title":"Neural networks with recurrent generative feedback","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"2020","author":"Huang"},{"key":"ref192","article-title":"Human eyes inspired recurrent neural networks are more robust against adversarial noises","author":"Choi","year":"2022","journal-title":"arXiv:2206.07282"},{"key":"ref193","doi-asserted-by":"publisher","DOI":"10.32470\/CCN.2019.1191-0"},{"key":"ref194","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr42600.2020.00950"},{"key":"ref195","doi-asserted-by":"publisher","DOI":"10.1162\/jocn_a_00282"},{"key":"ref196","first-page":"3910","article-title":"Adversarial examples that fool both computer vision and time-limited humans","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Elsayed"},{"key":"ref197","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i9.16936"},{"key":"ref198","article-title":"Liquid structural state-space models","volume-title":"11th Int. Conf. Learn. Represent","author":"Hasani"},{"key":"ref199","article-title":"Liquid neural network-based adaptive learning vs. incremental learning for link load prediction amid concept drift due to network failures","author":"Ayoub","year":"2024","journal-title":"arXiv:2404.05304"},{"key":"ref200","doi-asserted-by":"publisher","DOI":"10.1016\/j.conb.2020.11.009"},{"key":"ref201","first-page":"18132","article-title":"Adaptive recurrent vision performs zero-shot computation scaling to unseen difficulty levels","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"36","author":"Veerabadran"},{"key":"ref202","doi-asserted-by":"publisher","DOI":"10.1109\/3dv53792.2021.00032"},{"key":"ref203","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52688.2022.01578","article-title":"Practical stereo matching via cascaded recurrent network with adaptive correlation","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR)","author":"Li"},{"key":"ref204","doi-asserted-by":"crossref","DOI":"10.1109\/WACV57701.2024.00852","article-title":"Stereo matching in time: 100 FPS video stereo matching for extended reality","volume-title":"the IEEE\/CVF Winter Conf. Appl. Comput. Vis. (WACV)","author":"Cheng"},{"key":"ref205","doi-asserted-by":"publisher","DOI":"10.1109\/lra.2020.3017478"},{"key":"ref206","first-page":"1","article-title":"IamNN: Iterative and adaptive mobile neural network for efficient image classification","volume-title":"Proc. 6th Int. Conf. Learn. Represent. (ICLR)","author":"Leroux"},{"key":"ref207","first-page":"1","article-title":"Deep networks with internal selective attention through feedback connections","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Stollenga"},{"key":"ref208","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1011280"},{"key":"ref209","first-page":"5867","article-title":"Lifelong neural predictive coding: Learning cumulatively online without forgetting","volume-title":"Proc. 36th Int. Conf. Neural Inf. Process. Syst.","author":"Ororbia"},{"key":"ref210","article-title":"Reinforcement learning through active inference","author":"Tschantz","year":"2020","journal-title":"arXiv:2002.12636"},{"key":"ref211","article-title":"Deep predictive coding networks for video prediction and unsupervised learning","author":"Lotter","year":"2016","journal-title":"arXiv:1605.08104"},{"key":"ref212","doi-asserted-by":"publisher","DOI":"10.1162\/neco_a_01357"},{"key":"ref213","doi-asserted-by":"publisher","DOI":"10.1080\/17588928.2015.1020053"},{"key":"ref214","volume-title":"Reinforcement Learning with Prediction-based Rewards","year":"2024"},{"key":"ref215","article-title":"Active inference in robotics and artificial agents: Survey and challenges","author":"Lanillos","year":"2021","journal-title":"arXiv:2112.01871"},{"key":"ref216","article-title":"Predictive coding beyond correlations","volume-title":"41st Int. Conf. Mach. Learn","author":"Salvatori"},{"key":"ref217","article-title":"Neuronal circuit policies","author":"Lechner","year":"2018","journal-title":"arXiv:1803.08554"},{"key":"ref218","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-022-00556-7"},{"key":"ref219","article-title":"Worm-level control through search-based reinforcement learning","volume-title":"Deep Reinforcement Learn. Symp. 31st Neural Inf. Process. Syst. (NIPS)","author":"Lechner"},{"key":"ref220","article-title":"Interpretable recurrent neural networks in continuous-time control environments","author":"Hasani","year":"2020"},{"key":"ref221","doi-asserted-by":"publisher","DOI":"10.1523\/jneurosci.15-03-02434.1995"},{"key":"ref222","doi-asserted-by":"publisher","DOI":"10.1523\/jneurosci.05-04-00956.1985"},{"key":"ref223","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.0703183104"},{"key":"ref224","article-title":"SIM-CE: An advanced simulation platform for studying the brain of caenorhabditis elegans","author":"Hasani"},{"key":"ref225","volume-title":"Methods in Neuronal Modeling: From Ions to Networks","author":"Koch","year":"1998"},{"key":"ref226","article-title":"Can a compact neuronal circuit policy be re-purposed to learn simple robotic control?","author":"Hasani","year":"2019","journal-title":"arXiv:1809.04423"},{"key":"ref227","article-title":"Learning with chemical versus electrical synapses\u2014Does it make a difference?","author":"Farsang","year":"2023","journal-title":"arXiv:2401.08602"},{"key":"ref228","first-page":"4082","article-title":"A natural lottery ticket winner: Reinforcement learning with ordinary neural circuits","volume-title":"Proc. 37th Int. Conf. Mach. Learn. (PMLR)","author":"Hasani"},{"key":"ref229","article-title":"Liquid resistance liquid capacitance networks","volume-title":"1st Workshop NeurIPS","author":"Farsang"},{"key":"ref230","article-title":"On the benefits of biophysical synapses","author":"Lemmel","year":"2023","journal-title":"arXiv:2303.04944"},{"key":"ref231","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/w14-4012"},{"key":"ref232","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref233","doi-asserted-by":"publisher","DOI":"10.1016\/s0893-6080(05)80125-x"},{"key":"ref234","doi-asserted-by":"publisher","DOI":"10.1007\/s41870-023-01506-1"},{"key":"ref235","doi-asserted-by":"publisher","DOI":"10.3390\/electronics13142736"},{"key":"ref236","doi-asserted-by":"publisher","DOI":"10.1093\/pnasnexus\/pgae580"},{"key":"ref237","article-title":"End to end learning for self-driving cars","author":"Bojarski","year":"2016","journal-title":"arXiv:1604.07316"},{"key":"ref238","doi-asserted-by":"publisher","DOI":"10.1109\/lra.2022.3146555"},{"key":"ref239","article-title":"L46: Investigations into compressing closed form continuous time networks","author":"Agrawal"},{"key":"ref240","doi-asserted-by":"publisher","DOI":"10.3233\/faia240556"},{"key":"ref241","doi-asserted-by":"publisher","DOI":"10.1016\/s1053-8119(03)00202-7"},{"key":"ref242","volume-title":"MITCBMM, Liquid Neural Networks","year":"2021"},{"key":"ref243","first-page":"1310","article-title":"On the difficulty of training recurrent neural networks","volume-title":"Proc. 30th Int. Conf. Mach. Learn.","author":"Pascanu"},{"key":"ref244","article-title":"Learning long-term dependencies in irregularly-sampled time series","author":"Lechner","year":"2020","journal-title":"arXiv:2006.04418"},{"key":"ref245","doi-asserted-by":"publisher","DOI":"10.1109\/access.2025.3561593"},{"key":"ref246","article-title":"Efficiently modeling long sequences with structured state spaces","author":"Gu","year":"2021","journal-title":"arXiv:2111.00396"},{"key":"ref247","doi-asserted-by":"publisher","DOI":"10.1016\/s0959-4388(98)80042-1"},{"key":"ref248","doi-asserted-by":"publisher","DOI":"10.1093\/cercor\/1.1.1"},{"key":"ref249","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2012.272"},{"key":"ref250","doi-asserted-by":"publisher","DOI":"10.1007\/978-981-10-0213-7_4"},{"key":"ref251","doi-asserted-by":"publisher","DOI":"10.3389\/fncom.2014.00135"},{"key":"ref252","doi-asserted-by":"publisher","DOI":"10.1016\/s0896-6273(02)01091-7"},{"key":"ref253","doi-asserted-by":"publisher","DOI":"10.1037\/11496-001"},{"key":"ref254","doi-asserted-by":"publisher","DOI":"10.1017\/s0140525x23001644"},{"key":"ref255","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.4.721"},{"key":"ref256","doi-asserted-by":"publisher","DOI":"10.1162\/jocn_a_01544"},{"key":"ref257","doi-asserted-by":"publisher","DOI":"10.1007\/s11633-022-1348-x"},{"key":"ref258","article-title":"Are ViTs as global as we think?-Assessing model locality for brain-model mapping","author":"Huang"},{"key":"ref259","article-title":"On the relationship between self-attention and convolutional layers","volume-title":"8th Int. Conf. Learn. Represent. (ICLR)","author":"Cordonnier"},{"key":"ref260","doi-asserted-by":"publisher","DOI":"10.1109\/iccv.2017.89"},{"key":"ref261","first-page":"12116","article-title":"Do vision transformers see like convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Raghu"},{"key":"ref262","doi-asserted-by":"publisher","DOI":"10.3389\/fcomp.2023.1178450"},{"key":"ref263","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-psych-122414-033400"},{"key":"ref264","doi-asserted-by":"publisher","DOI":"10.1146\/annurev.neuro.23.1.315"},{"key":"ref265","doi-asserted-by":"publisher","DOI":"10.1016\/j.visres.2011.04.012"},{"key":"ref266","doi-asserted-by":"publisher","DOI":"10.1016\/j.cub.2018.04.041"},{"key":"ref267","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-30508-6_24"},{"key":"ref268","doi-asserted-by":"crossref","DOI":"10.1007\/978-981-96-0917-8_17","article-title":"Vision language models are blind","volume-title":"Proc. Asian Conf. Comput. Vis.","author":"Rahmanzadehgervi"},{"key":"ref269","volume-title":"Are We on the Right Way for Evaluating Large Vision-Language Models?","author":"Chen","year":"2024"},{"key":"ref270","first-page":"1","article-title":"Task-driven convolutional recurrent models of the visual system","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Nayebi"},{"key":"ref271","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyg.2013.00124"},{"key":"ref272","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuron.2014.06.017"},{"key":"ref273","article-title":"Seeing in the dark with recurrent convolutional neural networks","author":"Hartmann","year":"2018","journal-title":"arXiv:1811.08537"},{"key":"ref274","doi-asserted-by":"publisher","DOI":"10.32470\/ccn.2018.1152-0"},{"key":"ref275","article-title":"The lottery ticket hypothesis: Finding sparse, trainable neural networks","volume-title":"Int. Conf. Learn. Represent","author":"Frankle"},{"key":"ref276","doi-asserted-by":"publisher","DOI":"10.1214\/aos\/1176350051"},{"key":"ref277","first-page":"254","article-title":"Stronger generalization bounds for deep nets via a compression approach","volume-title":"Proc. 35th Int. Conf. Mach. Learn.","author":"Arora"},{"key":"ref278","article-title":"Non-vacuous generalization bounds at the Im-AGeNet scale: A pac-Bayesian compression approach","volume-title":"7th Int. Conf. Learn. Represent","author":"Zhou"},{"key":"ref279","article-title":"Residual connections encourage iterative inference","volume-title":"Int. Conf. Learn. Represent","author":"Jastrzebski"},{"key":"ref280","doi-asserted-by":"crossref","DOI":"10.2139\/ssrn.5080605","article-title":"Drawbacks of LSTM algorithm: A case study","author":"Kandadi","year":"2025"},{"key":"ref281","doi-asserted-by":"publisher","DOI":"10.1038\/4580"},{"key":"ref282","doi-asserted-by":"publisher","DOI":"10.1098\/rstb.2005.1622"},{"key":"ref283","article-title":"Predictive coding: A theoretical and experimental review","author":"Millidge","year":"2022","journal-title":"arXiv:2107.12979"},{"key":"ref284","doi-asserted-by":"publisher","DOI":"10.1038\/nrn2787"},{"key":"ref285","doi-asserted-by":"publisher","DOI":"10.1016\/j.bandc.2007.06.007"},{"key":"ref286","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2016.90","article-title":"Deep residual learning for image recognition","volume-title":"Proc. IEEE Conf. Comput. Vis. Pattern Recognit.","author":"He"},{"key":"ref287","doi-asserted-by":"publisher","DOI":"10.1109\/9780470544037.ch14"},{"key":"ref288","doi-asserted-by":"publisher","DOI":"10.1038\/337129a0"},{"key":"ref289","article-title":"Towards biologically plausible deep learning","author":"Bengio","year":"2016","journal-title":"arXiv:1502.04156"},{"key":"ref290","doi-asserted-by":"publisher","DOI":"10.1109\/5.58337"},{"key":"ref291","doi-asserted-by":"publisher","DOI":"10.1038\/s41583-020-0277-3"},{"key":"ref292","article-title":"Brain-inspired computational intelligence via predictive coding","author":"Salvatori","year":"2023","journal-title":"arXiv:2308.07870"},{"key":"ref293","doi-asserted-by":"publisher","DOI":"10.1162\/neco_a_01497"},{"key":"ref294","doi-asserted-by":"publisher","DOI":"10.1162\/NECO_a_00949"},{"key":"ref295","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2003.06.005"},{"key":"ref296","first-page":"38232","article-title":"Learning on arbitrary graph topologies via predictive coding","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Salvatori"},{"key":"ref297","doi-asserted-by":"publisher","DOI":"10.3389\/fncom.2022.1062678"},{"key":"ref298","doi-asserted-by":"publisher","DOI":"10.1038\/nrn.2017.149"},{"key":"ref299","first-page":"70548","article-title":"Neural foundations of mental simulation: Future prediction of latent representations on dynamic scenes","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Nayebi"},{"key":"ref300","article-title":"Joint embedding vs reconstruction: Provable benefits of latent space prediction for self supervised learning","author":"Van Assel","year":"2025","journal-title":"arXiv:2505.12477"},{"key":"ref301","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-022-29632-7"},{"key":"ref302","article-title":"Learning and leveraging world models in visual representation learning","author":"Garrido","year":"2024","journal-title":"arXiv:2403.00504"},{"key":"ref303","article-title":"Intuitive physics understanding emerges from self-supervised pretraining on natural videos","author":"Garrido","year":"2025","journal-title":"arXiv:2502.11831"},{"key":"ref304","article-title":"Higher-level spatial prediction in natural vision across mouse visual cortex","author":"Heilbron","year":"2025","journal-title":"BioRxiv"},{"issue":"1205","key":"ref305","first-page":"427","article-title":"Predictive coding: A fresh view of inhibition in the retina","volume":"216","author":"Srinivasan","year":"1997","journal-title":"Proc. Roy. Soc. London. Ser. B. Biol. Sci."},{"key":"ref306","article-title":"Meta-representational predictive coding: Biomimetic self-supervised learning","author":"Ororbia","year":"2025","journal-title":"arXiv:2503.21796"},{"key":"ref307","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-020-0170-9"},{"key":"ref308","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/12441.001.0001"},{"key":"ref309","doi-asserted-by":"publisher","DOI":"10.1007\/s00422-010-0364-z"},{"key":"ref310","doi-asserted-by":"publisher","DOI":"10.1016\/j.tics.2012.08.006"},{"key":"ref311","article-title":"Expected free energy-based planning as variational inference","author":"de Vries","year":"2025","journal-title":"arXiv:2504.14898"},{"key":"ref312","doi-asserted-by":"publisher","DOI":"10.3389\/fnbot.2021.651432"},{"key":"ref313","article-title":"Mastering diverse domains through world models","author":"Hafner","year":"2023","journal-title":"arXiv:2301.04104"},{"key":"ref314","article-title":"R-AIF: Solving sparse-reward robotic tasks from pixels with active inference and world models","author":"Nguyen","year":"2024","journal-title":"arXiv:2409.14216"},{"key":"ref315","volume-title":"Network-Based Active Inference for Adaptive and Cost-efficient Real-World Applications: A Benchmark Study of a Valve-Turning Task Against Deep Reinforcement Learning","author":"Singh","year":"2024"},{"key":"ref316","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2021.05.010"},{"key":"ref317","doi-asserted-by":"publisher","DOI":"10.1109\/icdl-epirob48136.2020.9278105"},{"key":"ref318","volume-title":"Towards Stable Learning in Predictive Coding Networks","author":"Ha","year":"2024"},{"key":"ref319","article-title":"Benchmarking predictive coding networks\u2014Made simple","volume-title":"13th Int. Conf. Learn. Represent","author":"Pinchetti"},{"key":"ref320","article-title":"\u03bcPC: Scaling predictive coding to 100+ layer networks","author":"Innocenti","year":"2025","journal-title":"arXiv:2505.13124"},{"key":"ref321","doi-asserted-by":"publisher","DOI":"10.1162\/neco_a_01620"},{"key":"ref322","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i10.28954"},{"key":"ref323","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2024.124315"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6287639\/10820123\/11080376.pdf?arnumber=11080376","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,26]],"date-time":"2025-07-26T06:32:05Z","timestamp":1753511525000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11080376\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":323,"URL":"https:\/\/doi.org\/10.1109\/access.2025.3589514","relation":{},"ISSN":["2169-3536"],"issn-type":[{"value":"2169-3536","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025]]}}}