Nature最新文献

{"title":"An integrated large-scale photonic accelerator with ultralow latency","authors":"Shiyue Hua, Erwan Divita, Shanshan Yu, Bo Peng, Charles Roques-Carmes, Zhan Su, Zhang Chen, Yanfei Bai, Jinghui Zou, Yunpeng Zhu, Yelong Xu, Cheng-kuan Lu, Yuemiao Di, Hui Chen, Lushan Jiang, Lijie Wang, Longwu Ou, Chaohong Zhang, Junjie Chen, Wen Zhang, Hongyan Zhu, Weijun Kuang, Long Wang, Huaiyu Meng, Maurice Steinman, Yichen Shen","doi":"10.1038/s41586-025-08786-6","DOIUrl":"10.1038/s41586-025-08786-6","url":null,"abstract":"Integrated photonics, particularly silicon photonics, have emerged as cutting-edge technology driven by promising applications such as short-reach communications, autonomous driving, biosensing and photonic computing1–4. As advances in AI lead to growing computing demands, photonic computing has gained considerable attention as an appealing candidate. Nonetheless, there are substantial technical challenges in the scaling up of integrated photonics systems to realize these advantages, such as ensuring consistent performance gains in upscaled integrated device clusters, establishing standard designs and verification processes for complex circuits, as well as packaging large-scale systems. These obstacles arise primarily because of the relative immaturity of integrated photonics manufacturing and the scarcity of advanced packaging solutions involving photonics. Here we report a large-scale integrated photonic accelerator comprising more than 16,000 photonic components. The accelerator is designed to deliver standard linear matrix multiply–accumulate (MAC) functions, enabling computing with high speed up to 1 GHz frequency and low latency as small as 3 ns per cycle. Logic, memory and control functions that support photonic matrix MAC operations were designed into a cointegrated electronics chip. To seamlessly integrate the electronics and photonics chips at the commercial scale, we have made use of an innovative 2.5D hybrid advanced packaging approach. Through the development of this accelerator system, we demonstrate an ultralow computation latency for heuristic solvers of computationally hard Ising problems whose performance greatly relies on the computing latency. A large-scale photonic accelerator comprising more than 16,000 components integrated on a single chip to process MAC operations is described, demonstrating ultralow latency and reduced computing time compared with a commercially available GPU.","PeriodicalId":18787,"journal":{"name":"Nature","volume":"640 8058","pages":"361-367"},"PeriodicalIF":50.5,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s41586-025-08786-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143806299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A map of neural signals and circuits traces the logic of brain computation","authors":"Mariela D. Petkova, \u0000 Gregor F. P. Schuhknecht","doi":"10.1038/d41586-025-00908-4","DOIUrl":"10.1038/d41586-025-00908-4","url":null,"abstract":"A large consortium of scientists adds functional data to neuronal wiring maps, seeking to crack the computational circuitry of the mouse brain. A large consortium of scientists adds functional data to neuronal wiring maps, seeking to crack the computational circuitry of the mouse brain.","PeriodicalId":18787,"journal":{"name":"Nature","volume":"640 8058","pages":"319-321"},"PeriodicalIF":50.5,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143806397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Foundation model of neural activity predicts response to new stimulus types","authors":"Eric Y. Wang, Paul G. Fahey, Zhuokun Ding, Stelios Papadopoulos, Kayla Ponder, Marissa A. Weis, Andersen Chang, Taliah Muhammad, Saumil Patel, Zhiwei Ding, Dat Tran, Jiakun Fu, Casey M. Schneider-Mizell, MICrONS Consortium, R. Clay Reid, Forrest Collman, Nuno Maçarico da Costa, Katrin Franke, Alexander S. Ecker, Jacob Reimer, Xaq Pitkow, Fabian H. Sinz, Andreas S. Tolias","doi":"10.1038/s41586-025-08829-y","DOIUrl":"10.1038/s41586-025-08829-y","url":null,"abstract":"The complexity of neural circuits makes it challenging to decipher the brain’s algorithms of intelligence. Recent breakthroughs in deep learning have produced models that accurately simulate brain activity, enhancing our understanding of the brain’s computational objectives and neural coding. However, it is difficult for such models to generalize beyond their training distribution, limiting their utility. The emergence of foundation models1 trained on vast datasets has introduced a new artificial intelligence paradigm with remarkable generalization capabilities. Here we collected large amounts of neural activity from visual cortices of multiple mice and trained a foundation model to accurately predict neuronal responses to arbitrary natural videos. This model generalized to new mice with minimal training and successfully predicted responses across various new stimulus domains, such as coherent motion and noise patterns. Beyond neural response prediction, the model also accurately predicted anatomical cell types, dendritic features and neuronal connectivity within the MICrONS functional connectomics dataset2. Our work is a crucial step towards building foundation models of the brain. As neuroscience accumulates larger, multimodal datasets, foundation models will reveal statistical regularities, enable rapid adaptation to new tasks and accelerate research. A foundation model trained on neural activity of visual cortex from multiple mice accurately predicts responses to video stimuli and cell types, dendritic features and connectivity within the MICrONS functional connectomics dataset.","PeriodicalId":18787,"journal":{"name":"Nature","volume":"640 8058","pages":"470-477"},"PeriodicalIF":50.5,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s41586-025-08829-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143806455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"New treatments to put insomnia to bed","authors":"","doi":"10.1038/d41586-025-00963-x","DOIUrl":"https://doi.org/10.1038/d41586-025-00963-x","url":null,"abstract":"Drugs that eliminate wakefulness, molecules in cannabis and wearable devices that modulate brain activity could help those with the condition.","PeriodicalId":18787,"journal":{"name":"Nature","volume":"24 1","pages":""},"PeriodicalIF":64.8,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143805632","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research round-up: sleep","authors":"","doi":"10.1038/d41586-025-00966-8","DOIUrl":"https://doi.org/10.1038/d41586-025-00966-8","url":null,"abstract":"The academic impact of early morning lectures, alcohol’s effects on sleep and other highlights from sleep studies.","PeriodicalId":18787,"journal":{"name":"Nature","volume":"108 1","pages":""},"PeriodicalIF":64.8,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143805634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Complete ape genomes offer a close-up view of human evolution","authors":"Lukas Kuderna","doi":"10.1038/d41586-025-00912-8","DOIUrl":"https://doi.org/10.1038/d41586-025-00912-8","url":null,"abstract":"Scientists have fully sequenced the genomes of six living ape species, enabling long-awaited comparisons of hard-to-assemble genomic regions.","PeriodicalId":18787,"journal":{"name":"Nature","volume":"21 1","pages":""},"PeriodicalIF":64.8,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143806123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Jurassic fossil reveals the origin of parasitic thorny-headed worms","authors":"","doi":"10.1038/d41586-025-01052-9","DOIUrl":"https://doi.org/10.1038/d41586-025-01052-9","url":null,"abstract":"A 165-million-year-old specimen clarifies the evolutionary history of a group of modern parasites.","PeriodicalId":18787,"journal":{"name":"Nature","volume":"183 1","pages":""},"PeriodicalIF":64.8,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143813619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Biggest brain map ever details huge number of neurons and their activity","authors":"","doi":"10.1038/d41586-025-01088-x","DOIUrl":"https://doi.org/10.1038/d41586-025-01088-x","url":null,"abstract":"3D reconstruction is the first to overlay neuronal activity on a large-scale map of brain cells.","PeriodicalId":18787,"journal":{"name":"Nature","volume":"59 1","pages":""},"PeriodicalIF":64.8,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143813887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"How do I explain the publication gap I ended up with after a hostile manager?","authors":"","doi":"10.1038/d41586-025-00982-8","DOIUrl":"https://doi.org/10.1038/d41586-025-00982-8","url":null,"abstract":"A poor working relationship with a supervisor or manager can result in lost opportunities, creating an unexplained gap in an early-career researcher’s publication record.","PeriodicalId":18787,"journal":{"name":"Nature","volume":"38 1","pages":""},"PeriodicalIF":64.8,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143805633","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Jurassic acanthocephalan illuminates the origin of thorny-headed worms","authors":"Cihang Luo, Luke A. Parry, Brendon E. Boudinot, Shengyu Wang, Edmund A. Jarzembowski, Haichun Zhang, Bo Wang","doi":"10.1038/s41586-025-08830-5","DOIUrl":"https://doi.org/10.1038/s41586-025-08830-5","url":null,"abstract":"<p>Acanthocephala (thorny-headed worms), characterized by the presence of an eversible proboscis with hooks, are a diverse endoparasitic group that infect a wide range of vertebrates and invertebrates¹. Although long regarded as a separate phylum, they have several putative sister taxa based on morphological features, including Platyhelminthes (flatworms)², Priapulida (penis worms)³ and Rotifera (wheel animals)⁴. Molecular phylogenies have instead recovered them within rotifers^5,6,7,8,9,10, suggesting acanthocephalans are derived from free-living worms with a jaw apparatus (Gnathifera). Their only fossil record is Late Cretaceous eggs¹¹, contributing limited palaeontological information to deciphering their early evolution. Here we describe an acanthocephalan body fossil, <i>Juracanthocephalus daohugouensis</i> gen. et. sp. nov., from the Middle Jurassic Daohugou biota of China. <i>Juracanthocephalus</i> shows unambiguous acanthocephalan characteristics, for example a hooked proboscis, a bursa, as well as a jaw apparatus with discrete elements that is typical of other gnathiferans. <i>Juracanthocephalus</i> shares features with Seisonidea (an epizoic member of Rotifera) and Acanthocephala, bridging the evolutionary gap between jawed rotifers and the obligate parasitic, jawless acanthocephalans. Our results reveal previously unrecognized ecological and morphological diversity in ancient Acanthocephala and highlight the significance of transitional fossils, revealing the origins of this highly enigmatic group of living organisms.</p>","PeriodicalId":18787,"journal":{"name":"Nature","volume":"7 1","pages":""},"PeriodicalIF":64.8,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143806193","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}