{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T04:23:40Z","timestamp":1783052620676,"version":"3.54.6"},"reference-count":47,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,12,1]],"date-time":"2026-12-01T00:00:00Z","timestamp":1796083200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/100004358","name":"Samsung Electronics Co Ltd","doi-asserted-by":"publisher","award":["IO201210-08019-01"],"award-info":[{"award-number":["IO201210-08019-01"]}],"id":[{"id":"10.13039\/100004358","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":[[2026,12]]},"DOI":"10.1016\/j.neunet.2026.109222","type":"journal-article","created":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T23:49:37Z","timestamp":1780616977000},"page":"109222","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["Temperature-free loss function for contrastive learning"],"prefix":"10.1016","volume":"204","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4155-9225","authenticated-orcid":false,"given":"Bum Jun","family":"Kim","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6023-1837","authenticated-orcid":false,"given":"Sang Woo","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.neunet.2026.109222_bib0001","series-title":"ICML","first-page":"589","article-title":"Adaptive sampled softmax with kernel based sampling","volume":"vol. 80","author":"Blanc","year":"2018"},{"key":"10.1016\/j.neunet.2026.109222_bib0002","series-title":"NeurIPS","article-title":"Unsupervised learning of visual features by contrasting cluster assignments","author":"Caron","year":"2020"},{"key":"10.1016\/j.neunet.2026.109222_bib0003","series-title":"ICML","first-page":"1597","article-title":"A simple framework for contrastive learning of visual representations","volume":"vol. 119","author":"Chen","year":"2020"},{"key":"10.1016\/j.neunet.2026.109222_bib0004","series-title":"KDD","first-page":"767","article-title":"On sampling strategies for neural network-based collaborative filtering","author":"Chen","year":"2017"},{"issue":"9","key":"10.1016\/j.neunet.2026.109222_bib0005","doi-asserted-by":"crossref","first-page":"1734","DOI":"10.1109\/TPAMI.2015.2496141","article-title":"Discriminative unsupervised feature learning with exemplar convolutional neural networks","volume":"38","author":"Dosovitskiy","year":"2016","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"10.1016\/j.neunet.2026.109222_bib0006","series-title":"ICML","first-page":"1436","article-title":"BOHB: Robust and efficient hyperparameter optimization at scale","volume":"vol. 80","author":"Falkner","year":"2018"},{"key":"10.1016\/j.neunet.2026.109222_bib0007","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1016\/j.neunet.2022.11.009","article-title":"DANet: Semi-supervised differentiated auxiliaries guided network for video action recognition","volume":"158","author":"Gao","year":"2023","journal-title":"Neural Networks"},{"key":"10.1016\/j.neunet.2026.109222_bib0008","series-title":"ACM DL","first-page":"89","article-title":"CiteSeer: An automatic citation indexing system","author":"Giles","year":"1998"},{"key":"10.1016\/j.neunet.2026.109222_bib0009","series-title":"NeurIPS","article-title":"Bootstrap your own latent - a new approach to self-supervised learning","author":"Grill","year":"2020"},{"issue":"4","key":"10.1016\/j.neunet.2026.109222_bib0010","doi-asserted-by":"crossref","first-page":"19:1","DOI":"10.1145\/2827872","article-title":"The MovieLens datasets: History and context","volume":"5","author":"Harper","year":"2016","journal-title":"ACM Transactions on Interactive Intelligent Systems"},{"key":"10.1016\/j.neunet.2026.109222_bib0011","series-title":"EMNLP","first-page":"9681","article-title":"MABEL: Attenuating gender bias using textual entailment data","author":"He","year":"2022"},{"key":"10.1016\/j.neunet.2026.109222_bib0012","series-title":"CVPR","first-page":"9726","article-title":"Momentum contrast for unsupervised visual representation learning","author":"He","year":"2020"},{"key":"10.1016\/j.neunet.2026.109222_bib0013","series-title":"CVPR","first-page":"770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"key":"10.1016\/j.neunet.2026.109222_bib0014","series-title":"NIPS","first-page":"1731","article-title":"Train longer, generalize better: Closing the generalization gap in large batch training of neural networks","author":"Hoffer","year":"2017"},{"key":"10.1016\/j.neunet.2026.109222_bib0015","series-title":"Applied logistic regression, second edition","author":"Hosmer","year":"2000"},{"key":"10.1016\/j.neunet.2026.109222_bib0016","unstructured":"Howard, J. (2019). Imagenette: A smaller subset of 10 easily classified classes from Imagenet. https:\/\/github.com\/fastai\/imagenette."},{"key":"10.1016\/j.neunet.2026.109222_bib0017","doi-asserted-by":"crossref","first-page":"313","DOI":"10.1016\/j.neunet.2023.07.010","article-title":"Perceptual contrastive generative adversarial network based on image warping for unsupervised image-to-image translation","volume":"166","author":"Huang","year":"2023","journal-title":"Neural Networks"},{"key":"10.1016\/j.neunet.2026.109222_bib0018","doi-asserted-by":"crossref","DOI":"10.1016\/j.neunet.2024.106517","article-title":"DPGCL: Dual pass filtering based graph contrastive learning","volume":"179","author":"Huang","year":"2024","journal-title":"Neural Networks"},{"key":"10.1016\/j.neunet.2026.109222_bib0019","series-title":"ICML","first-page":"13774","article-title":"Model-aware contrastive learning: Towards escaping the dilemmas","volume":"vol. 202","author":"Huang","year":"2023"},{"key":"10.1016\/j.neunet.2026.109222_bib0020","series-title":"ECCV (7)","first-page":"67","article-title":"Learning visual features from large weakly supervised data","volume":"vol. 9911","author":"Joulin","year":"2016"},{"key":"10.1016\/j.neunet.2026.109222_bib0021","series-title":"ICLR","article-title":"On large-batch training for deep learning: generalization gap and sharp minima","author":"Keskar","year":"2017"},{"key":"10.1016\/j.neunet.2026.109222_bib0022","series-title":"ICLR","article-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2015"},{"key":"10.1016\/j.neunet.2026.109222_bib0023","unstructured":"Krizhevsky, A., Hinton, G. et al. (2009). Learning multiple layers of features from tiny images."},{"key":"10.1016\/j.neunet.2026.109222_bib0024","series-title":"ICLR","article-title":"Temperature schedules for self-supervised contrastive methods on long-tail data","author":"Kukleva","year":"2023"},{"key":"10.1016\/j.neunet.2026.109222_bib0025","series-title":"NeurIPS","first-page":"6423","article-title":"On controllable sparse alternatives to softmax","author":"Laha","year":"2018"},{"key":"10.1016\/j.neunet.2026.109222_bib0026","doi-asserted-by":"crossref","DOI":"10.1016\/j.neunet.2024.106584","article-title":"Towards a rigorous analysis of mutual information in contrastive learning","volume":"179","author":"Lee","year":"2024","journal-title":"Neural Networks"},{"key":"10.1016\/j.neunet.2026.109222_bib0027","series-title":"ICLR","article-title":"SGDR: Stochastic gradient descent with warm restarts","author":"Loshchilov","year":"2017"},{"key":"10.1016\/j.neunet.2026.109222_bib0028","article-title":"DySTreSS: Dynamically scaled temperature in self-supervised contrastive learning","volume":"abs\/2308.01140","author":"Manna","year":"2023","journal-title":"CoRR"},{"key":"10.1016\/j.neunet.2026.109222_bib0029","series-title":"ACL\/IJCNLP (1)","first-page":"5356","article-title":"StereoSet: Measuring stereotypical bias in pretrained language models","author":"Nadeem","year":"2021"},{"key":"10.1016\/j.neunet.2026.109222_bib0030","series-title":"EMNLP (1)","first-page":"1953","article-title":"CrowS-Pairs: A challenge dataset for measuring social biases in masked language models","author":"Nangia","year":"2020"},{"key":"10.1016\/j.neunet.2026.109222_sbref0031","article-title":"Representation learning with contrastive predictive coding","author":"van den Oord","year":"2018","journal-title":"CoRR"},{"key":"10.1016\/j.neunet.2026.109222_bib0032","series-title":"GECCO companion","first-page":"241","article-title":"Learning multiple defaults for machine learning algorithms","author":"Pfisterer","year":"2021"},{"key":"10.1016\/j.neunet.2026.109222_bib0033","series-title":"ICML","first-page":"28389","article-title":"Not all semantics are created equal: Contrastive self-supervised learning with automatic temperature individualization","volume":"vol. 202","author":"Qiu","year":"2023"},{"key":"10.1016\/j.neunet.2026.109222_bib0034","series-title":"ICML","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume":"vol. 139","author":"Radford","year":"2021"},{"key":"10.1016\/j.neunet.2026.109222_bib0035","series-title":"AAAI","first-page":"2155","article-title":"Mean-shifted contrastive loss for anomaly detection","author":"Reiss","year":"2023"},{"key":"10.1016\/j.neunet.2026.109222_bib0036","series-title":"NIPS","first-page":"1849","article-title":"Improved deep metric learning with multi-class N-pair loss objective","author":"Sohn","year":"2016"},{"key":"10.1016\/j.neunet.2026.109222_bib0037","series-title":"CVPR","first-page":"2495","article-title":"Understanding the behaviour of contrastive loss","author":"Wang","year":"2021"},{"key":"10.1016\/j.neunet.2026.109222_bib0038","series-title":"CVPR","first-page":"3733","article-title":"Unsupervised feature learning via non-parametric instance discrimination","author":"Wu","year":"2018"},{"key":"10.1016\/j.neunet.2026.109222_bib0039","series-title":"NeurIPS","first-page":"17084","article-title":"Tuning large neural networks via zero-shot hyperparameter transfer","author":"Yang","year":"2021"},{"key":"10.1016\/j.neunet.2026.109222_bib0040","series-title":"WWW","first-page":"1063","article-title":"Debiased contrastive learning for sequential recommendation","author":"Yang","year":"2023"},{"key":"10.1016\/j.neunet.2026.109222_bib0041","series-title":"ECCV (26)","first-page":"668","article-title":"Decoupled contrastive learning","volume":"vol. 13686","author":"Yeh","year":"2022"},{"key":"10.1016\/j.neunet.2026.109222_bib0042","series-title":"CVPR","first-page":"14421","article-title":"Dual temperature helps contrastive learning without many negative samples: Towards understanding and simplifying moco","author":"Zhang","year":"2022"},{"key":"10.1016\/j.neunet.2026.109222_bib0043","series-title":"ICLR","article-title":"How does SimSiam avoid collapse without negative samples? A unified understanding with self-supervised contrastive learning","author":"Zhang","year":"2022"},{"key":"10.1016\/j.neunet.2026.109222_sbref0044","article-title":"Temperature as uncertainty in contrastive learning","author":"Zhang","year":"2021","journal-title":"CoRR"},{"key":"10.1016\/j.neunet.2026.109222_bib0045","doi-asserted-by":"crossref","DOI":"10.1016\/j.neunet.2024.106557","article-title":"DECNet: Dense embedding contrast for unsupervised semantic segmentation","volume":"179","author":"Zhang","year":"2024","journal-title":"Neural Networks"},{"key":"10.1016\/j.neunet.2026.109222_bib0046","series-title":"NeurIPS datasets and benchmarks","article-title":"An empirical study of graph contrastive learning","author":"Zhu","year":"2021"},{"key":"10.1016\/j.neunet.2026.109222_sbref0047","article-title":"Deep graph contrastive representation learning","author":"Zhu","year":"2020","journal-title":"CoRR"}],"container-title":["Neural Networks"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0893608026006830?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0893608026006830?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T04:10:12Z","timestamp":1783051812000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0893608026006830"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,12]]},"references-count":47,"alternative-id":["S0893608026006830"],"URL":"https:\/\/doi.org\/10.1016\/j.neunet.2026.109222","relation":{},"ISSN":["0893-6080"],"issn-type":[{"value":"0893-6080","type":"print"}],"subject":[],"published":{"date-parts":[[2026,12]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Temperature-free loss function for contrastive learning","name":"articletitle","label":"Article Title"},{"value":"Neural Networks","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.neunet.2026.109222","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":"109222"}}