{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,2]],"date-time":"2026-11-02T08:27:44Z","timestamp":1793608064838,"version":"4.5.0"},"reference-count":43,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2027,2,1]],"date-time":"2027-02-01T00:00:00Z","timestamp":1801440000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2027,2,1]],"date-time":"2027-02-01T00:00:00Z","timestamp":1801440000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2027,2,1]],"date-time":"2027-02-01T00:00:00Z","timestamp":1801440000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2027,2,1]],"date-time":"2027-02-01T00:00:00Z","timestamp":1801440000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2027,2,1]],"date-time":"2027-02-01T00:00:00Z","timestamp":1801440000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2027,2,1]],"date-time":"2027-02-01T00:00:00Z","timestamp":1801440000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2027,2,1]],"date-time":"2027-02-01T00:00:00Z","timestamp":1801440000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62471411"],"award-info":[{"award-number":["62471411"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Neural Networks"],"published-print":{"date-parts":[[2027,2]]},"DOI":"10.1016\/j.neunet.2026.109720","type":"journal-article","created":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T15:27:22Z","timestamp":1791473242000},"page":"109720","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"PA","title":["Accelerating training speed of tiny recursive models with curriculum guided adaptive recursion"],"prefix":"10.1016","volume":"206","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0102-3816","authenticated-orcid":false,"given":"Kaleem Ullah","family":"Qasim","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiashu","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4296-3510","authenticated-orcid":false,"given":"Razan","family":"Alharith","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad Kafeel","family":"Shaheen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.neunet.2026.109720_bib0001","unstructured":"Banino, A., Balaguer, J., & Blundell, C. (2021). Pondernet: Learning to ponder. arXiv: 2107.05407."},{"key":"10.1016\/j.neunet.2026.109720_bib0002","series-title":"Proceedings of the 26th annual international conference on machine learning","first-page":"41","article-title":"Curriculum learning","author":"Bengio","year":"2009"},{"key":"10.1016\/j.neunet.2026.109720_bib0003","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109720_bib0004","series-title":"Advances in neural information processing systems","first-page":"6571","article-title":"Neural ordinary differential equations","volume":"vol. 31","author":"Chen","year":"2018"},{"key":"10.1016\/j.neunet.2026.109720_bib0005","series-title":"International conference on machine learning","first-page":"794","article-title":"Gradnorm: Gradient normalization for adaptive loss balancing in deep multitask networks","author":"Chen","year":"2018"},{"key":"10.1016\/j.neunet.2026.109720_bib0006","unstructured":"Chollet, F. (2019). On the measure of intelligence. arXiv: 1911.01547."},{"issue":"240","key":"10.1016\/j.neunet.2026.109720_bib0007","first-page":"1","article-title":"PaLM: Scaling language modeling with pathways","volume":"24","author":"Chowdhery","year":"2023","journal-title":"Journal of Machine Learning Research"},{"key":"10.1016\/j.neunet.2026.109720_bib0008","series-title":"International conference on learning representations (ICLR)","article-title":"Universal transformers","author":"Dehghani","year":"2019"},{"key":"10.1016\/j.neunet.2026.109720_bib0009","unstructured":"Graves, A. (2016). Adaptive computation time for recurrent neural networks. arXiv: 1603.08983."},{"key":"10.1016\/j.neunet.2026.109720_bib0010","series-title":"International conference on machine learning","first-page":"1311","article-title":"Automated curriculum learning for neural networks","author":"Graves","year":"2017"},{"key":"10.1016\/j.neunet.2026.109720_bib0011","series-title":"Proceedings of the 37th international conference on machine learning","first-page":"3822","article-title":"Breaking the curse of space explosion: Towards efficient NAS with curriculum search","volume":"vol. 119","author":"Guo","year":"2020"},{"key":"10.1016\/j.neunet.2026.109720_bib0012","series-title":"International conference on machine learning","first-page":"2535","article-title":"On the power of curriculum learning in training deep networks","author":"Hacohen","year":"2019"},{"issue":"11","key":"10.1016\/j.neunet.2026.109720_bib0013","doi-asserted-by":"crossref","first-page":"7436","DOI":"10.1109\/TPAMI.2021.3117837","article-title":"Dynamic neural networks: A survey","volume":"44","author":"Han","year":"2022","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"10.1016\/j.neunet.2026.109720_bib0014","series-title":"International conference on learning representations (ICLR)","article-title":"LoRA: Low-rank adaptation of large language models","author":"Hu","year":"2022"},{"key":"10.1016\/j.neunet.2026.109720_bib0015","series-title":"European conference on computer vision","first-page":"646","article-title":"Deep networks with stochastic depth","author":"Huang","year":"2016"},{"key":"10.1016\/j.neunet.2026.109720_bib0016","series-title":"Ecai 2024 - 27th European conference on artificial intelligence","article-title":"CBM: Curriculum by masking","author":"Jarc\u02c7a","year":"2024"},{"key":"10.1016\/j.neunet.2026.109720_bib0017","doi-asserted-by":"crossref","unstructured":"Jolicoeur-Martineau, A. (2025). Less is more: Recursive reasoning with tiny networks. arXiv: 2510.04871.","DOI":"10.21203\/rs.3.rs-8148771\/v1"},{"key":"10.1016\/j.neunet.2026.109720_bib0018","series-title":"Advances in neural information processing systems","article-title":"Why warmup the learning rate? Underlying mechanisms and improvements","volume":"vol. 37","author":"Kalra","year":"2024"},{"key":"10.1016\/j.neunet.2026.109720_bib0019","unstructured":"Koishekenov, Y., Lipani, A., & Cancedda, N. (2025). Encode, think, decode: Scaling test-time reasoning with recursive latent thoughts. arXiv: 2510.07358."},{"key":"10.1016\/j.neunet.2026.109720_bib0020","series-title":"Artificial intelligence and statistics","first-page":"562","article-title":"Deeply-supervised nets","author":"Lee","year":"2015"},{"key":"10.1016\/j.neunet.2026.109720_bib0021","series-title":"Proceedings of the AAAI conference on artificial intelligence","first-page":"1504","article-title":"Curriculum temperature for knowledge distillation","volume":"vol. 37","author":"Li","year":"2023"},{"key":"10.1016\/j.neunet.2026.109720_bib0022","series-title":"International conference on learning representations (ICLR)","article-title":"Decoupled weight decay regularization","author":"Loshchilov","year":"2019"},{"key":"10.1016\/j.neunet.2026.109720_bib0023","doi-asserted-by":"crossref","first-page":"46534","DOI":"10.52202\/075280-2019","article-title":"Self-refine: Iterative refinement with self-feedback","volume":"36","author":"Madaan","year":"2023","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109720_bib0024","unstructured":"Ng, K.-S., & Wang, Q. (2024). Loop neural networks for parameter sharing. arXiv: 2409.14199."},{"key":"10.1016\/j.neunet.2026.109720_bib0025","doi-asserted-by":"crossref","DOI":"10.1613\/jair.1.18562","article-title":"Recursive decomposition of logical thoughts: Framework for superior reasoning and knowledge propagation in large language models","volume":"83","author":"Qasim","year":"2025","journal-title":"Journal of Artificial Intelligence Research"},{"key":"10.1016\/j.neunet.2026.109720_bib0026","series-title":"Proceedings of the 2024 joint international conference on computational linguistics, language resources and evaluation (LREC-COLING 2024)","first-page":"5212","article-title":"Does the order matter? Curriculum learning over languages","author":"Ranaldi","year":"2024"},{"key":"10.1016\/j.neunet.2026.109720_bib0027","series-title":"Proceedings of the 14th international conference on recent advances in natural language processing","first-page":"937","article-title":"Modeling easiness for training transformers with curriculum learning","author":"Ranaldi","year":"2023"},{"key":"10.1016\/j.neunet.2026.109720_bib0028","unstructured":"Raposo, D., Ritter, S., Richards, B., Lillicrap, T., Humphreys, P. C., & Santoro, A. (2024). Mixture-of-depths: Dynamically allocating compute in transformer-based language models. arXiv: 2404.02258."},{"key":"10.1016\/j.neunet.2026.109720_bib0029","unstructured":"Rusu, A. A., Rabinowitz, N. C., Desjardins, G., Soyer, H., Kirkpatrick, J., Kavukcuoglu, K., Pascanu, R., & Hadsell, R. (2016). Progressive neural networks. arXiv: 1606.04671."},{"key":"10.1016\/j.neunet.2026.109720_bib0030","series-title":"Proceedings of the 2025 conference of the nations of the americas chapter of the association for computational linguistics: Human language technologies (volume 1: Long papers)","first-page":"8418","article-title":"Thread: Thinking deeper with recursive spawning","author":"Schroeder","year":"2025"},{"key":"10.1016\/j.neunet.2026.109720_bib0031","doi-asserted-by":"crossref","first-page":"8634","DOI":"10.52202\/075280-0377","article-title":"Reflexion: Language agents with verbal reinforcement learning","volume":"36","author":"Shinn","year":"2023","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109720_bib0032","series-title":"International conference on learning representations (ICLR)","article-title":"Scaling LLM test-time compute optimally can be more effective than scaling model parameters","author":"Snell","year":"2025"},{"issue":"6","key":"10.1016\/j.neunet.2026.109720_bib0033","doi-asserted-by":"crossref","first-page":"1526","DOI":"10.1007\/s11263-022-01611-x","article-title":"Curriculum learning: A survey","volume":"130","author":"Soviany","year":"2022","journal-title":"International Journal of Computer Vision"},{"key":"10.1016\/j.neunet.2026.109720_bib0034","series-title":"International conference on machine learning","first-page":"10096","article-title":"Efficientnetv2: Smaller models and faster training","author":"Tan","year":"2021"},{"issue":"6","key":"10.1016\/j.neunet.2026.109720_bib0035","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3530811","article-title":"Efficient transformers: A survey","volume":"55","author":"Tay","year":"2022","journal-title":"ACM Computing Surveys"},{"issue":"9","key":"10.1016\/j.neunet.2026.109720_bib0036","doi-asserted-by":"crossref","first-page":"12288","DOI":"10.1109\/TNNLS.2023.3255783","article-title":"Improving deep neural networks\u2019 training for image classification with nonlinear conjugate gradient-style adaptive momentum","volume":"35","author":"Wang","year":"2024","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"10.1016\/j.neunet.2026.109720_bib0037","unstructured":"Wang, G., Li, J., Sun, Y., Chen, X., Liu, C., Wu, Y., Lu, M., Song, S., & Yadkori, Y. A. (2025). Hierarchical reasoning model. arXiv: 2506.21734."},{"key":"10.1016\/j.neunet.2026.109720_bib0038","series-title":"European conference on computer vision","first-page":"420","article-title":"Skipnet: Learning dynamic routing in convolutional networks","author":"Wang","year":"2018"},{"key":"10.1016\/j.neunet.2026.109720_bib0039","series-title":"Proceedings of the 16th conference of the european chapter of the association for computational linguistics","first-page":"91","article-title":"Berxit: Early exiting for bert with better fine-tuning and extension to regression","author":"Xin","year":"2021"},{"key":"10.1016\/j.neunet.2026.109720_bib0040","series-title":"IEEE conference on computer vision and pattern recognition","first-page":"675","article-title":"Pad-net: Multi-tasks guided prediction-and-distillation network for simultaneous depth estimation and scene parsing","author":"Xu","year":"2018"},{"issue":"8","key":"10.1016\/j.neunet.2026.109720_bib0041","doi-asserted-by":"crossref","first-page":"15029","DOI":"10.1109\/TNNLS.2025.3544724","article-title":"Spcnet: Deep self-paced curriculum network incorporated with inductive bias","volume":"36","author":"Zhao","year":"2025","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"10.1016\/j.neunet.2026.109720_bib0042","series-title":"Proceedings of the 30th ACM international conference on multimedia","first-page":"6792","article-title":"Curriculum-NAS: Curriculum weight-sharing neural architecture search","author":"Zhou","year":"2022"},{"key":"10.1016\/j.neunet.2026.109720_bib0043","series-title":"Proceedings of the thirty-second international joint conference on artificial intelligence","first-page":"6823","article-title":"A survey on efficient training of transformers","author":"Zhuang","year":"2023"}],"container-title":["Neural Networks"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0893608026011755?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0893608026011755?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,11,2]],"date-time":"2026-11-02T07:46:13Z","timestamp":1793605573000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0893608026011755"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2027,2]]},"references-count":43,"alternative-id":["S0893608026011755"],"URL":"https:\/\/doi.org\/10.1016\/j.neunet.2026.109720","relation":{},"ISSN":["0893-6080"],"issn-type":[{"value":"0893-6080","type":"print"}],"subject":[],"published":{"date-parts":[[2027,2]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Accelerating training speed of tiny recursive models with curriculum guided adaptive recursion","name":"articletitle","label":"Article Title"},{"value":"Neural Networks","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.neunet.2026.109720","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"109720"}}