1,022 | 0 | 52 |
下载次数 | 被引频次 | 阅读次数 |
以ChatGPT为代表的大语言模型的发展标志人工智能(AI)进入“通用人工智能”发展的新时代。综述了通用人工智能“大数据、小任务”专用人工智能发展阶段的两大热点:人工智能大语言模型和AI芯片的最新进展和发展趋势。在人工智能大语言模型领域,综述并分析了其发展由来和发展现状,包括专家系统和聊天机器人两条技术路线的发展历程,OpenAI的ChatGPT领跑大模型的发展现状,以及对大模型的综述、深化、改进并推向应用的新进展。在AI芯片领域,综述并分析了在人工智能大模型发展带动下,云计算AI芯片和边缘计算AI芯片的最新进展,包括新一代GPU、TPU、云计算AI芯片新架构、NPU架构的边缘计算AI芯片、数字边缘计算AI芯片、数字CIM基模拟AI芯片和模拟CIM AI芯片。大语言模型创新涌现的特点和AI芯片架构创新的黄金时代特征应该值得高度关注。
Abstract:The development of large language models represented by ChatGPT marks that artificial intelligence(AI) has entered a new era of “artificial general intelligence” development. Two major hotspots in the development of specialized artificial intelligence from “big data, small tasks” in the era of general artificial intelligence are reviewed: the latest progress and trends in artificial intelligence large language models and AI chips. In the field of artificial intelligence large language models, the origins and current status of development are reviewed and analyzed, including the development trajectories of expert systems and chatbots, OpenAI's ChatGPT leads the way in the development of large models, and the new progress in summarizing, deepening, improving, and pushing forward the application of large models. In the field of AI chips, the latest progress in cloud computing AI chips and edge computing AI chips driven by the development of large language models are reviewed and analyzed, including the new generation of GPUs, TPUs, the new architecture of cloud computing AI chips, NPU architectures for edge computing AI chips, digital edge AI chips, digital CIM-based analog AI chips, and analog CIM AI chips. It is worth noting that the features of the emergence of large language model innovation and the golden age of AI chip architecture innovation.
[1]WANG F Y,MIAO Q H,LI X,et al.What does ChatGPTsay:the DAO from algorithmic intelligence to linguistic intelligence[J].CAA Journal of Automatica Sinica,2023,10(3):575-579.
[2]Chat GPT问世,AI赛道关注度有望持续提升[EB/OL].(2023-02-07)[2024-09-15]https://xueqiu.com/1597430632/241482130.
[3]LIANG S.AI computing in large-scale era:pre-trillionscale neural network models andexa-scale supercomputing[C]//Proceedings of International VLSI Symposium on Technology,Systems and Applications (VLSI-TSA/VLSI-DAT).Hsin Cuh,China,2023:10134466
[4]LENAT D,PRAKASH M,SHEPHERD M.CYC:using common sense know-ledge to overcome brittleness and knowledge acquisition bottlenecks[J].AI Magazine,1985,6(4):65-85
[5]FERRUCCI D,LEVAS A,BAGCHI S,et al.Watson:beyond Jeopardy![J].Artificial Intelligence,2013,199/200:93-105.
[6]VASWANI A,SHAZEER N,PARMAR N,et.al.Attention is all you need[C]//Proceedings of the 31st Conference on Neural Information Processing Systems.Long Beach,USA,2017:6000-6010.
[7]BROWN B,MANN B,RYDER N,et al.Language models are few-shot learners[C]//Proceedings of the 34th Conference on Neural Information Processing Systems (Neur ISP2020) Vancouver,Canada,2020:2005.14165.
[8]GPT-3.5语言模型:AI领域的最新热点[EB/OL].(2023-04-18)[2024-09-15].https://baijiahao.baidu.com/s?id=1763477539919029599&wfr=spider&for=pc.
[9]MIAO H,ZHENG W B,LV Y S,et al.DAO to HANOIvia De Sic:AI paradigm shifts from Alpha Go to Chat GPT[J].IEEE/CAA Journal of Automatica Sinica,2023,10(4):877-897.
[10]聊天机器人的革命:Open AI推出了基于GPT-4模型的Chat GPT平台[EB/OL].(2023-03-15)[2024-09-15]https://www.bilibili.com/read/cv22410204/.
[11]界面新闻.Open AI宣布将GPT-4引入内容审核系统,减少人工参与[EB/OL].(2023-08-16)[2024-09-15]https://t.cj.sina.com.cn/articles/view/5182171545/134e1a99902001lrcp.
[12]Open AI.Video generation models as world simulators[EB/OL].(2024-02-15)[2024-10-15]https://openai.com/index/video-generation-models-as-world-simulators/.
[13]Open AI.Introducing Open AI o1-preview[EB/OL].(2024-09-12)[2024-10-15]https://openai.com/index/introducing-openai-o1-preview/.
[14]BERGHEL H.ChatGPT and AIChat epistemology[J].Computer,2023,56 (5):130-137.
[15]GRUDIN J.Chat GPT and chat history:challenges for the new wave[J].Computer,2023,56 (5):94-100.
[16]WANG Y,LI JJ,QIN R,et al.Chat GPT for computational social systems:from conversational applications to human-oriented operating systems[J].IEEE Transactions on Computational Social Systems,2023,10 (2):414-425.
[17]ABDULLAH M,MADAIN A,JARARWEH Y.ChatGPT:fundamentals,applications and social impacts[C]//Proceedings of Ninth International Conference on Social Networks Analysis,Management and Security (SNAMS).Milan,Italy,2022:10062688.
[18]PANCHBHAI A,PANKANTI S.Exploring large language models in limited resource scenario[C]//Proceedings of the 11th International Conference on Cloud Computing,Data Science Engineering (Confluence).Noida,India,2021:147-152.
[19]HASHANA M J,BRUNDHA P,AYOOBKHAN M UA,et al.Deep learning in Cat Gh PT:survey[C]//Proceedings of the 7th International Conference on Trends in Electronics and Informatics (ICOEI).Tirunelveli,India,2023:1001-1005.
[20]MANODNYA H,GIRI A.GPT-K:a GPT-based model for generation of text in Kannada[C]//Proceedings of IEEE 4th International Conference on Cybernetics,Cognition and Machine Learning Applications (ICCCMLA).Goa,India,2022:534-539.
[21]CHEN M,TWOREK J,JUN H,et al.Evaluating large language models trained on code[J/OL].(2021-07-14)[2024-10-15]https://arxiv.org/abs/2107.03374.
[22]AUSTIN J,ODENA A,NYE M,et al.Program synthesis with large language models[J/OL].(2021-08-16)[2024-10-15]https://arxiv.org/abs/2108.07732.
[23]新智元.UC伯克利LLM排行榜首次重磅更新!GPT-4稳居榜首,全新330亿参数“小羊驼”位列开源第一[EB/OL].(2023-06-23)[2024-9-20]https://zhuanlan.zhihu.com/p/607403006.
[24]百度.什么是最适合中国的大模型:新华网评测:文心一言得分最高领先GPT-3.5[EB/OL].(2023-08-04)[2024-10-15]https://baijiahao.baidu.com/s?id=1773282791051587211&wfr=spider&for=pc.
[25]C?MARA V,MENDONCA-NETO R,SILVA A,et al.large language model approach to SQL-to-text generation[C]//Proceedings of IEEE International Conference on Consumer Electronics (ICCE).Las Vegas,USA,2024:10444148
[26]WANG Y F,GUO Q Y,NI X Z,et al.Hint-enhanced incontext learning wakes largelanguage models up for knowledge-intensive tasks[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:10276-10280.
[27]NAM L,PARK J,CHOI J,et al.Language-oriented communication with semantic coding and knowledge distillation for text-to-image generation[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:13506-13510.
[28]LONG J,KILLICK G,MCCREADIE R,et al.Multiway-adapter:adapting multimodal large language models for scalable image-text retrieval[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:6580-6584.
[29]HWANG Y,LEE J,PARK J,et al.Searching optimal floating-point format for sub-8-bit large language model inference[C]//Proceedings of International Conference on Electronics,Information,and Communication (ICEIC).Taipei,China,2024:1-4
[30]HIGUCHI Y,OGAWA T,KOBAYASHI T,et al.BEC-TRA:transducer-basedend-to-end ASR with bert-enhanced encoder[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (IC-ASSP).Rhodes Island,Greece,2023:10095186.
[31]CHEN Z,ALLAUZEN C,HUANG Y H,et al.Largescale language model rescoring on long-form data[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Rhodes Island,Greece,2023:10096429.
[32]OGAWA A,TAWARA N,DELCROIX M,et al.Lattice rescoring based on large ensemble of complementary neural language models[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (IC-ASSP).Singapore,Singapore,2022:6517-6521.
[33]DAS N,SUNKARA M,BODAPATI S,et al.Mask the bias:improving domain-adaptive generalization of CTC-based ASR with internal language model estimation[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Rhodes Island,Greece,2023:10096405.
[34]KUBO Y,KARITA S,BACCHIANI M.Knowledge transfer from large-scalepretrained language models to end-to-end speech recognizers[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (IC-ASSP).Singapore,2022:8512-8516.
[35]MADDIGAN P,SUSNJAK T.Chat2VIS:generating data visualizations via natural language using ChatGPT,codex and GPT-3 large language models[J].IEEE Access,2023,11:45181-45193.
[36]STROBELT H,WEBSON A,SANH V,et al.Interactive and visual prompt engineering for ad-hoc task adaptation with large language models[J].IEEE Transactions on Visualization and Computer Graphics,2023,29 (1):1146-1156.
[37]JAIN N,VAIDYANATH S,IYER A,et al.Jigsaw:large language models meet program synthesis[C]//Proceedings of the 44th International Conference on Software Engineering.Pittsburgh,USA,2022:1219-1231.
[38]HUANG R,ALLAUZEN C,CHEN T Z,et al.Multilingual and fully non-autoregressive ASR with large language model fusion:comprehensive study[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:13306-13310.
[39]UDAGAWA T,SUZUKI M,KURATA G,et al.Multiple representation transfer from large language models to endto-end ASR systems[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:10176-10180.
[40]ADAMSON G.Explaining technology we do not understand[J].IEEE Transactions on Technology and Society,2023,4 (1):34-45.
[41]BEHNIA R,EBRAHIMI R,PACHECO J,et al.EW-tune:framework for privately fine-tuning large language models with differential privacy[C]//Proceedings of IEEEInternational Conference on Data Mining Workshops (ICD-MW).Orlando,USA,2022:560-566.
[42]HAJI S,SUEKANE K,SANO H,et al.Exploratory inference chain:exploratorily chaining multi-hop inferences with large language models for question-answering[C]//Proceedings of IEEE 17th International Conference on Semantic Computing (ICSC).Laguna Hills,USA,2023:175-182.
[43]ZHU R,ZHANG M.How robust is a large pre-trained language model for code generation?case on attacking GPT2[C]//Proceedings of IEEE International Conference on Software Analysis,Evolution and Reengineering (SAN-ER).Taipa,China,2023:708-712.
[44]YAO Y,ZHANG J S,HARRIS I G,et al.Fuzz LLM:a novel and universal fuzzing framework for proactively discovering jailbreak vulnerabilities in large language models[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:4485-4489.
[45]KHOJE M.Navigating data privacy and analytics:the role of large language models in masking conversational data in data platforms[C]//Proceedings of IEEE 3rd International Conference on AI in Cybersecurity (ICAIC).Houston,USA,2024:10433801.
[46]PATWARY M.Keynotetalk 2 training large language models:challenges and opportunities[C]//Proceedings of IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW).Lyon,France,2022:1245.
[47]VERES C.Large language models are not models of natural language:they are corpus models[J].IEEE Access,2022,10:61970-61979.
[48]MA Z,MA J B,LIU X J,et al.Large margin training for long short-term memory neural networks in neural language modeling[C]//Proceedings of the 5th International Conference on Pattern Recognition and Artificial Intelligence (PRAI).Chengdu,China,2022:673-677.
[49]WANG L,HUANG JJ,CHURCH K W.Large margin training improves language models for ASR[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Toronto,Canada,2021:7368-7372.
[50]HU K,SAINATH N,LI B,et al.Massively multilingual shallow fusion with large language models[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Rhodes Island,Greece,2023:1-5.
[51]WU S,CHEN K,ZHANG T Y,et al.Large-scale contrastive language-audiopretraining with feature fusion and keyword-to-caption augmentation[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Rhodes Island,Greece,2023:1-5.
[52]PANDELEA V,RAGUSA E,GASTALDO P,et al.Selecting language models features VIA software-hardware codesign[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Rhodes Island,Greece,2023:1-5.
[53]BERRY L,SHIH J,WANG H F,et al.M-SpeechCLIP:leveraging large-scale,pre-trained models for multilingual speech to image retrieval[C]//Proceedings of IEEEInternational Conference on Acoustics,Speech and Signal Processing (ICASSP).Rhodes Island,Greece,2023:1-5.
[54]SAMMANI F,MUKHERJEE T,DELIGIANNIS N.NLX-GPT:model for natural language explanations in vision and vision-language tasks[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).New Orleans,USA,2022:8312-8322.
[55]MOUKAFIH Y,GHOGHO M,SMAILI K.Supervised contrastive learning as multi-objective optimization for finetuning large pre-trained language models[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Pocessinr g (ICASSP).Rhodes Island,Greece,2023:1-5.
[56]XU B,SONG C Z,TIAN Y,et al.Training largevocabulary neural language models by private federated learning for resource-constrained devices[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Rhodes Island,Greece,2023:1-5.
[57]逆天“魔法”正式解禁!GPT-4以来最强应用“代码解释器”上线![EB/OL].(2023-07-09)[2024-09-22]https://baijiahao.baidu.com/s?id=1770924122176777043&wfr=spider&for=pc.
[58]Open AI反水微软!Altman密谋“私人订制Chat GPT”,AI市场的未来又要变天?[EB/OL].(2023-07-13)[2024-09-22]https://www.51cto.com/article/760476.html.
[59]引发新一轮技术革命的AIGC,市场潜力有多大[EB/OL].(2023-07-13)[2024-09-22]https://baijiahao.baidu.com/s?id=1771310008222549907&wfr=spider&for=pc.
[60]国君计算机:AI+办公是此次AIGC浪潮中的核心受益方[EB/OL].(2023-07-11)[2024-09-22].https://stock.jrj.com.cn/2023/07/11071737682049.shtml.
[61]PRASANNA R,RITHANI M,BHARATHI M G,et al.Empirical evaluation of large language models in resume classification[C]//Proceedings of Fourth International Conference on Advances in Electrical,Computing,Communication and Sustainable Technologies (ICAECT).Bhilai,India,2024:1-4.
[62]HAN W,LU J K,XU Y,et al.Intelligent practices of large language models in digital government services[J].IEEE Access,2024,12:8633-8640.
[63]LUO C,WANG J S,ZHOU A J,et al.Large language models augmented rating prediction in recommender system[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:7960-7964.
[64]KODALI K,UPRETI Y P,BOPPANA L.Large language models in AWS[C]//Proceedings of 1st International Conference on Robotics,Engineering,Science,and Technology (RESTCON).Pattaya,Thailand,2024:112-117.
[65]ZHANG P,PU J,XUE J R,et al.Hi Ve GPT:humanmachine-augmented intelligent vehicles with generative pretrained transformer[J].IEEE Transactions on Intelligent Vehicles,2023,8 (3):2027-2033.
[66]DU P,TENG S Y,CHEN H,et al.Chat with Chat GPTon intelligent vehicles:an IEEE TIV perspective[J].IEEETransactions on Intelligent Vehicles,2023,8 (3):2020-2026.
[67]GAO B,TONG W,WU E Q,et al.Chat with Chat GPTon interactive engines for intelligent driving[J].IEEETransactions on Intelligent Vehicles,2023,8 (3):2034-2036.
[68]WANG Y.Linguistic intelligence for intelligent vehicles:Chat GPT and future logistics and mobility[J].IEEETransactions on Intelligent Vehicles,2023,8 (3):2011-2019.
[69]TANG Q,DAI X Y,ZHAO C,et al.Large language model-driven urban traffic signal control[C]//Proceedings of Australian New Zealand Control Conference (ANZCC).Gold Coast,Australia,2024:67-71.
[70]TANAKA Y,KATSURA S.voice-controlled motion reproduction using large language models for polishing robots[C]//Proceedings of IEEE International Conference on Mechatronics (ICM).Loughborough,UK,2023:1-6.
[71]MATHUR A,PRADHAN S,SONI P,et al.Automated test case generation using T5 and GPT-3[C]//Proceedings of the 9th International Conference on Advanced Computing and Communication Systems (ICACCS).Coimbatore,India,2023:1986-1992.
[72]WANG Y,YANG J,WANG XX,et al.Chat with ChatGPT on industry 5.0:learning and decision-making for intelligent industries[J].CAA Journal of Automatica Sinica,2023,10 (4):831-834.
[73]TSIGKANOS C,RANI P,MüLLER S,et al.Large language models:the next frontier for variable discovery within metamorphic testing?[C]//Proceedings of IEEE International Conference on Software Analysis,Evolution and Reengineering (SANER).Taipa,China,2023:678-682.
[74]ARNAUTOV V,AKIMOV D A.Application of large language models for optimization of electric power system states[C]//Proceedings of Conference of Young Researchers in Electrical and Electronic Engineering (El Cno).Saint Petersburg,Russian,2024:314-317.
[75]AI和量子技术的突破:预示着科技爆发和人类未来的新纪元[EB/OL].(2024-08-10)[2024-09-22]https://baijiahao.baidu.com/s?id=1806976477105845786&wfr=spider&for=pc.
[76]百度这波赢麻了!文心大模型3.5扒掉了所有国产AI大模型的“底裤”[EB/OL].(2023-07-21)[2024-09-22].https://baijiahao.baidu.com/s?id=1772031607161182282&wfr=spider&for=pc.
[77]RALTE Z,DAS R,KAR I.Deepcheque:large language model approach to automated Cheque collection framework and information retrival using multiple loss functions[C]//Proceedings of the 9th International Conference on Advanced Computing and Communication Systems (ICACCS).Coimbatore,India,2023:1786-1793.
[78]TOPSAKAL O,SAWYER P,AKINCI C,et al.Algorithms to measure area and volume on 3 face models for facial surgeries[J].IEEE Access,2023,11:39577-39585.
[79]KIRUTHIGA S.Doctormate-an early disease prediction approach using multiple machine learning algorithms[C]//Proceedings of Second International Conference on Electronics and Renewable Systems (ICEARS).Tuticorin,India,2023:1276-1282.
[80]KALLA A,MUKHOPADHYAY S,RALTE Z,et al.Exploring the impact of motif-driven causal temporal analysis using graph neural network in improving large language model performance forpharmacovigilance[C]//Proceedings of the 9th International Conference on Advanced Computing and Communication Systems (ICACCS).Coimbatore,India,2023:1769-1776.
[81]KUZLU M,XIAO X,SARP S,et al.The rise of generative artificial intelligence in Healthcare[C]//Proceedings of the 12th Mediterranean Conference on Embedded Computing (MECO).Budva,Montenegro,2023:1-4.
[82]XU Z.Using large pre-trained language model to assist FDA in premarket medical device classification[C]//Proceedings of Southeast Cn.o Orlando,USA,2023:159-166.
[83]PULA R,NEPACINA R,ILAGAN L.Classification of short-interfering RNA through transformer encoder model[C]//Proceedings of the 1st International Conference on Robotics,Engineering,Science,and Technology (REST-CON).Pattaya,Thailand,2024:191-194.
[84]SINGH A,EHTESHAM A,MAHMUD S,et al.Revolutionizing mental health care through Lang Chain:journey with large language model[C]//Proceedings of IEEE14th Annual Computing and Communication Workshop and Conference (CCWC).Las Vegas,USA,2024:73-78.
[85]GAO Y,ZHANG Y W,CHEN Y Q,et al.Unsupervised human activity recognition via large language models and iterative evolution[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (IC-ASSP).Seoul,Korea,2024:91-95.
[86]华为盘古大模型3.0正式发布!一句对话生成代码,能解决世界难题[EB/OL].(2023-07-10)[2024-09-22]https://www.360kuai.com/pc/93d29cb983cde1373?cota=3&kuai_so=1&sign=360_57c3bbd1&refer_scene=so_1.
[87]GUO C,LU Y,DOU Y,et al.Can Chat GPT boost artistic creation:the need of imaginative intelligence for parallel art[J].IEEE/CAA Journal of Automatica Sinica,2023,10(4):835-838.
[88]AMIN M,CAMBRIA E,SCHULLER B W.Will affective computing emerge from foundation models and general artificial intelligence?first evaluation of Chat GPT[J].IEEE Intelligent Systems,2023,38 (2):15-23.
[89]GAO J.Intelligent model description language and algorithm generation based on deep learning[C]//Proceedings of IEEE International Conference on Control,Electronics and Computer Technology (ICCECT).Jilin,China,2023:1333-1337.
[90]SHOUFAN A.Exploring students’perceptions of Chat GPT:thematic analysis and follow-up survey[J].IEEE Access,2023,11:38805-38818.
[91]IBRAHIM H,ASIM R,ZAFFAR F,et al.Rethinking homework in the age of artificial intelligence[J].IEEE Intelligent Systems,2023,38 (2):24-27.
[92]KOVAˇCEVIC'D.Use of Chat GPT in ESP teaching process[C]//Proceedings of the 22nd International Symposium Infoteh-Jahorina (INFOTEH).East Sarajevo,Bosnia and Herzegovina,2023:1-5.
[93]XU B,FU J.A two-stage sign language recognition method focusing on the semantic features of label text[C]//Proceedings of the 20th CSI International Symposium on Artificial Intelligence and Signal Processing (AISP).Babol,Iran,2024:1-5.
[94]HUANG Q,HUANG F,TAO H,et al.Co Q:an empirical framework for multi-hop question answering empowered by large language models[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:11566-11570.
[95]MISHRA M,BRAHAM A,MARSOM C,et al.DataAgent:evaluating large language models’ability to answer zero-shot,natural language queries[C]//Proceedings of IEEE 3rd International Conference on AI in Cybersecurity (ICAIC).Houston,USA,2024:1-5.
[96]MAHMUD A I,TALHA TALUKDER A A,SULTA-NA A,et al.Toward news authenticity:synthesizing natural language processing and human expert opinion to evaluate news[J].IEEE Access,2023,11:11405-11421.
[97]ONCESCU M,HENRIQUES J F,ZISSERMAN A,et al.sound approach:using large language models to generate audio descriptions for egocentric text-audio retrieval[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:7300-7304.
[98]GUAN F,KITAYAMA D.Review prediction using largescale language models for serendipity-oriented tourist spot recommendation and its evaluation[C]//Proceedings of the 18th International Conference on Ubiquitous Information Management and Communication (IMCOM).Kuala Lumpur,Malaysia,2024:1-4.
[99]EVERSON K,GU L,YANG H,et al.Towards ASRrobust spoken language understanding through in-context learning with word confusion networks[C]//Proceedings of IEEE International Conference on Acoustics,Speech and Signal Processing (ICASSP).Seoul,Korea,2024:12856-12860.
[100]SORU T,MARSHALL J.Trend extraction and analysis via large language models[C]//Proceedings of IEEE 18th International Conference on Semantic Computing (ICSC).Laguna Hills,USA,2024:285-288.
[101]新华网.法媒文章:美国人工智能巨头盯上国防领域[EB/OL].(2023-08-04)[2024-09-25].http://www.xinhuanet.com/mil/2023-08/04/c_1212252002.htm.
[102]GRBIC V,DUJLOVIC I.Social engineering with ChatGPT[C]//Proceedings of the 22nd International Symposium Infoteh-Jahorina (INFOTEH).East Sarajevo,Bosnia and Herzegovina,2023:1-5.
[103]澎湃新闻.AI四巨头将成立行业机构,规范人工智能发[EB/OL].(2023-07-27)[2024-09-25].https://baijiahao.baidu.com/s?id=1772552477757853334&wfr=spider&for=pc.
[104]英伟达推出AI芯片Blackwell GPU,成本和能耗降低25倍[EB/OL].(2024-03-19)[2024-09-25].https://zhuanlan.zhihu.com/p/687932189.
[105]谷歌发布第五代TPU:训练性能提高2倍,推理性能提升2.5倍![EB/OL].(2023-08-30)[2024-09-25]https://baijiahao.baidu.com/s?id=1775657695054416967&wfr=spider&for=pc.
[106]AMD MI300加速器深度揭秘:八路并行破亿亿次!全面超越NVIDIA[EB/OL].(2023-12-07)[2024-09-25].https://baijiahao.baidu.com/s?id=1784583621449965393&wfr=spider&for=pc.
[107]AI专题:AI加速计算需求,台积电ISSCC展望先进制程和先进封装新技术[EB/OL].(2024-03-07)[2024-09-25]https://www.sohu.com/a/762540393_121822496
[108]MOON S,KIM J,KIM H,et al.Hyper Acelc latency processing unit (LPUTM) accelerating hyperscale models for generative AI[C]//Proceedings of IEEE Hot Chips35 Symposium (HCS).Palo Alto,USA,2023:1.
[109]LIE S.Inside the Crebrase wafer-scale cluster[C]//Proceedings of IEEE Hot Chips 35 Symposium (HCS).Palo Alto,USA,2023:1-41.
[110]KWON Y,KIM G,KIM N,et al.Memory-centric computing with SK hynix’s domain-specific memory[C]//Proceedings of IEEE Hot Chips 35 Symposium (HCS).Palo Alto,USA,2023:1-26.
[111]XIAO B.Moffettantoum?:a deep-sparse AI inference system-on-chip for vision and large-language models[C]//Proceedings of IEEE Hot Chips 35 Symposium (HCS).Palo Alto,USA,2023:1-33.
[112]MAHURIN E.Qualocmm?HexagonTM NPU[C]//Proceedings of IEEE Hot Chips 35 Symposium (HCS).Palo Alto,USA,2023:1-19.
[113]TU B,WANG Y Q,WU Z H,et al.Re DCIM:reconfigurable digital computing-in-memory processor with unified FP/INT pipeline for cloud AI acceleration[J].IEEEJournal of Solid-State Circuits,2023,58 (1):243-255.
[114]XUE Y,WEN M,CHEN Z Y,et al.Releasing the potential of tensor core for unstructured Sp MM using tiledCSR format[C]//Proceedings of IEEE 41st International Conference on Computer Design (ICCD).Washington,USA,2023:457-464.
[115]KIM H,RO Y W,SO J N,et al.Samsung PIM/PNMfor transformer based AI:energy efficiency on PIM/PNMcluster[C]//Proceedings of IEEE Hot Chips 35 Symposium (HCS).Palo Alto,USA,2023,10254711:1-31.
[116]SPRACKLEN L,AHMAD S.Supercharged AI inference on modern CPUs[C]//Proceedings of IEEE Hot Chips35 Symposium (HCS).Palo Alto,USA,2023:1-21.
[117]YU H,KIM H E,SHIN S,et al.2.4 ATOMUS:a5 nm 32 TFLOPS/128 TOPS ML system-on-chip for latency critical applications[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,USA,2024:42-44.
[118]KAR M,SILBERMAN J,VENKATARAMANI S,et al.14.1 software-assisted peak current regulation scheme to improve power-limited inference performance in 5 nm AI-So C[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,USA,2024:254-256.
[119]与GPU双向奔赴,NPU即将开启大规模商用时代![EB/OL].(2024-03-09)[2024-09-25]https://www.eefocus.com/article/1672400.html.
[120]CONTI F,ROSSI D,PAULIN G,et al.22.112.4 TOPS/W@136 GOPS AI-IoTs ystem-on-chip with16 RISC-V,2-to-8b precision-scalable DNN acceleration and 30%-boost adaptive body biasing[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISS-CC).San Francisco,USA,2023:21-23
[121]ZHANG L,HUO D X,ZHANG J,et al.22.6 ANP-I:28 nm 1.5 p J/SOP asynchronous spiking neural network processor enabling sub-0.1μJ/sample on-chip learning for edge-AI applications[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,USA,2023:21-23
[122]LI Z,HSU Y C,SUMIKAWA R,et al.A 0.13 m J/prediction CIFAR-100 raster-scan-based wired-logic processor using non-linear neural network[C]//Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS).Monterey,USA,2023:1-5
[123]FANG M,WANG C Q,ZHAO S Q,et al.A 510μW0.738-mm 26.2-p J/SOP online learning multi-topology SNN processor with unified computation engine in 40-nm CMOS[J].IEEE Transactions on Biomedical Circuits and Systems,2023,17 (3):507-520
[124]PAPANDREOU N,van LUNTEREN J,ANGHEL A,et al.Acceleration of decision-tree ensemble models on the IBM telum processor[C]//Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS).Monterey,USA,2023:1-5
[125]GAO M,MOSANU S,SAKIB M N,et al.Lite ARI5:a system-level framework for the design and modeling of AI-extended RISC-V cores[C]//Proceedings of IEEE 36th International System-on-Chip Conference (SOCC).Santa Clara,USA,2023:1-6
[126]KAMBLE S,KULKARNI A.Memory utilization of AIapplication for DSP and ARMprocessor[C]//Proceedings of the 7th International Conference On Computing,Communication,Control and Automation (ICCUBEA).Pune,India,2023:1-4.
[127]GOMONY D,de PUTTER F,GEBREGIORGIS A,et al.Peta Osp/W edge-AI processors:myth or reality?[C]//Proceedings of Design,Automation Test in Europe Conference Exhibition (DATE).Antwerp,Belgium,2023:1-6
[128]ZHANG L,HUO D X,ZHANG J,et al.ANP-I:a28-nm 1.5-p J/SOP asynchronous spiking neural network processor enabling sub-0.1-μJ/sample on-chip learning for edge-AI applications[J].IEEE Journal of Solid-State Circuits,2024,59 (8):2717-2729
[129]LIANG Y,CHANG Y R,YANG P H,et al.A high efficiency hardware accelerator for convolution neural network[C]//Proceedings of the 9th International Conference on Applied System Innovation (ICASI).Chiba,Japan,2023:157-159
[130]HAN D,RYU J,KIM S,et al.low-power neural 3Drendering processor with bio-inspired visual perception core and hybrid DNN acceleration[C]//Proceedings of IEEESymposium in Low-Power and High-Speed Chips (COOLCHIPS).Tokyo,Japan,2023:1-3
[131]CHEN S,LI X Y,LU J C,et al.A reusable AI acceleration architecture based on matrix multiplication for convolutional neural network with digital signal processing tasks[C]//Proceedings of IEEE 15th International Conference on ASIC (ASICON).Nanjing,China,2023:1-4
[132]HU H,LIU X J,LIU Y,et al.A tiny accelerator for mixed-bit sparse CNN based on efficient fetch method of SIMO SPad[J].IEEE Transactions on Circuits and Systems II:Express Briefs,2023,70 (8):3079-3083
[133]HUANG S,CHOU T H,LU J M,et al.Hier Achr:a cluster-based DNN accelerator with hierarchical buses for design space exploration[C]//Proceedings of IEEE 36th International System-on-Chip Conference (SOCC).Santa Clara,USA,2023:1-6
[134]GILBERT M,WU N,PARASHAR A,et al.Loop Tree:enabling exploration of fused-layer dataflow accelerators[C]//Proceedings of IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS).Raleigh,USA,2023:316-318
[135]VARMA H C,SWAMINATHAN A,LIN S Y.Multimode AI accelerator architecture for thermal-aware 3stacked deep neural network design[C]//Proceedings of International Conference on Consumer Electronics Taiwan (ICCE-Taiwan).Ping Tung,China,2023:637-638
[136]BOK K,LEE S,KIM A,et al.Real-time inference platform for object detection on edge device[C]//Proceedings of International Technical Conference on Circuits/Systems,Computers,and Communications (ITC-CSCC).Jeju,Korea,2023:1-4
[137]SUZUKI J,YU J,YASUNAGA M,et al.Pianissimo:sub-m Wclass DNN accelerator with progressively adjustable bit-precision[J].IEEE Access,2023,12:2057-2073
[138]BAI R.A flexible and low-resource CNN accelerator on FPGA for edge computing[C]//Proceedings of the 3rd International Conference on Neural Networks,Information and Communication Engineering (NNICE).Guangzhou,China,2023:646-650
[139]WANG T,ZHAO L T,WU W,et al.Dynamic neural network accelerator for multispectral detection based on FPGA[C]//Proceedings of the 25th International Conference on Advanced Communications Technology(ICACT).Pyeongchang,Korea,2023:345-350.
[140]ROY A,KAPILA V,GUPTA A,et al.novel network on chip architecture for FPGA smart NIC[C]//Proceedings of IEEE Women in Technology Conference (WINTE-CHCON).Bangalore,India,2023:1-5.
[141]JEONG R,CHO K,JEONG Y,et al.A real-time reconfigurable AI processor based on FPGA[C]//Proceedings of IEEE International Conference on Consumer E-lectronics (ICCE).Las Vegas,USA,2023:1-2.
[142]LEFTHERIOTIS A,TZOMAKA A,DANOPOULOS D,et al.Evaluating Versal ACAP and conventional FPGAplatforms for AI inference[C]//Proceedings of the 12th International Conference on Modern Circuits and Systems Technologies (MOCAST).Athens,Greece,2023:1-6
[143]BHUSHAN P,GURUPRASADH J P,SIDHANATHS.FPGA based acceleration of a custom deep neural network model inference[C]//Proceedings of International Conference on Integration of Computational Intelligent System (ICICIS).Pune,India,2023:1-5
[144]LI Y.FPGA theoretical analysis and its advantage comparison in artificial intelligence[C]//Proceedings of IEEEInternational Conference on Image Processing and Computer Applications (ICIPCA).Changchun,China,2023:1885-1888
[145]SOLA-THOMAS E,BASET SARKER A,IMTIAZM.FPGA-controlled AI vision for prosthetics hand[C]//Proceedings of IEEE World AI Io T Congress (AIIo T).Seattle,USA,2023:520-524.
[146]ALHUSSAIN A,LIN J.FPGA-QHAR:throughputoptimized for quantized two-stream human action recognition on the edge[C]//Proceedings of IEEE 20th International Conference on Smart Communities:Improving Quality of Life Using AI,Robotics and Io T (HONET).Boca Raton,USA,2023:156-160
[147]ZYLIN'SKI M,NASSIBI A,RAKHMATULIN I,et al.Deployment of artificial intelligence models on edge devices:tutorial brief[J].IEEE Transactions on Circuits and Systems II:Express Briefs,2024,71 (3):1738-1743
[148]DONG C,JIA T Y,DU K X,et al.A model-specific end-to-end design methodology for resource-constrained Tiny ML hardware[C]//Proceedings of the 60th ACM/IEEE Design Automation Conference (DAC).San Francisco,USA,2023:1-6
[149]van DELM J,VANDERSTEEGEN M,BURRELLO A,et al.HTVM:efficient neural network deployment on heterogeneous Tiny ML platforms[C]//Proceedings of the 60th ACM/IEEE Design Automation Conference (DAC).San Francisco,USA,2023:1-6
[150]DING C,GU Y,DU Y,et al.A reconfigurable 2D-mesh No Cdesi gn with agile development technique of SpinalHDL[C]//Proceedings of International Symposium of Electronics Design Automation,Nanjing,China,2023:142-145.
[151]SKRBEK M,KUBALíK P,KOHLíK M,et al.Evaluation of the medium-sized neural network using approximative computations on zynq FPGA[C]//Proceedings of the12th Mediterranean Conference on Embedded Computing (MECO).Budva,Montenegro,2023:1-4
[152]WULFERT L,KUHNEL J,KRUPP L,et al.AIf ES:next-generation edge AI framework[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2024,46 (6):4519-4533
[153]LI Y,MA S C,WANG T Y,et al.HASP:hierarchical asynchronous parallelism for multi-NN tasks[J].IEEETransactions on Computers,2024,73 (2):366-379.
[154]HUANG T,LUO T,YAN M,et al.RCT:resource constrained training for edge AI[J].IEEE Transactions on Neural Networks and Learning Systems,2024,35 (2):2575-2587
[155]TAN H,WU Y F,ZHANG Y N,et al.A scalable multi-chiplet deep learning accelerator with hub-side 2.5heterogeneous integration[C]//Proceedings of IEEEHot Chips 35 Symposium (HCS).Palo Alto,USA,2023:1-17.
[156]KWON Y,HAN J,CHO P,et al.Chiplet hetero geneous-integration AI processor[C]//Proceedings of International Conference on Electronics,Information,and Communication (ICEIC).Singapore,2023:1-2
[157]STEVENS A,PAN T H,RAVICHANDIRAN P P,et al.Chiplet set for artificial intelligence[C]//Proceedings of IEEE International 3 Systems Integration Conference (3DIC).Cork,Ireland,2023:1-5
[158]HO D,JAMES SU Y,PU J,et al.Chiplet solution with FO-MCM package in edge and cloud computing (IMPACT2023)[C]//Proceedings of the 18th International Microsystems,Packaging,Assembly and Circuits Technology Conference (IMPACT).Taipei,China,2023:42-45.
[159]LIN C,ZHOU X Y,LIU R K,et al.Compare with the traditional heterogeneous solution:accelerate neural network algorithm through heterogeneous integrated CPUNPU chip on server[C]//Proceedings of IEEE 3rd International Conference on Computer Communication and Artificial Intelligence (CCAI).Taiyuan,China,2023:45-49.
[160]LU L,CHEN C C,LIN S C,et al.Demonstration of A3 chip by logic-DRAM stacked using paired TSV interconnection through interface for AI/edge-computing application[C]//Proceedings of International VLSI Symposium on Technology,Systems and Applications (VLSI-TSA/VLSI-DAT).Hsin Cuh,China,2023:1-2
[161]FEKETE G,KOVáCSHáZY T.Execution of resource intensive tasks on heterogeneous So Cfor low-latency embedded compute[C]//Proceedings of the 24th International Carpathian Control Conference (ICCC).MiskolcSzilvásvárad,Hungary,2023:124-129
[162]SAKUMA K,BONILLA G,MCHERRON D,et al.Heterogeneous Integration on Organic Interposer Substrate with fine-pitch RDL and 40 micron pitch micro-bumps[C]//Proceedings of IEEE 73rd Electronic Components and Technology Conference (ECTC).Orlando,USA,2023:872-877
[163]ITURBE X,ABDERRAHMANE N,ABELLA J,et al.Nimble AI:towards neuromorphic sensing-processing 3D-integrated chips[C]//Proceedings of Design,Automation Test in Europe Conference&Exhibition (DATE).Antwerp,Belgium,2023:1-6.
[164]SHE R,FU J.Research on evaluation strategy of heterogeneous computing chip based on improved common origin grey clustering[C]//Proceedings of the 4th International Conference on Computer Engineering and Application (ICCEA).Hangzhou,China,2023:65-69
[165]NOSE K,FUJII T,TOGAWA K,et al.20.323.9 TOPS/W@0.8 V,130 TOPS AI accelerator with16×performance-accelerable pruning in 14 nm heterogeneous embedded MPU for real-time robot applications[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,USA,2024:364-366
[166]CHEN Y,WU M,ZHAO W T,et al.7.8 A 22 nm delta-sigma computing-In-memory (Δ∑CIM) SRAM macro with near-zero-mean outputs and LSB-first ADCs achieving21.38 TOPS/W for 8 b-MAC edge AI processing[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,USA,2023:140-142
[167]TU B,WANG Y Q,WU Z H,et al.16.4 Tensor CMI:28 nm 3.7 n J/gather and 8.3 TFLOPS/W FP32digital-CIM tensor processor for MCM-CIM-based beyondNN acceleration[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,USA,2023:254-256
[168]MAI Z,WANG M Y,ZHANG C H,et al.A1.97 TFLOPS/W configurable SRAM-based floating-point computation-in-memory macro for energy-efficient AI chips[C]//Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS).Monterey,USA,2023:1-5
[169]WU C,SU J W,HONG L Y,et al.A 22 nm 832 kb hybrid-domain floating-point SRAM in-memory-compute macro with 16.2-70.2 TFLOPS/W for high-accuracy AI-edge devices[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,USA,2023:126-128.
[170]ALI M,CHAKRABORTY I,CHOUDHARY S,et al.65 nm 1.4-6.7 TOPS/W adaptive-SNR sparsity-aware CIM core with load balancing support for DL workloads[C]//Proceedings of IEEE Custom Integrated Circuits Conference (CICC).San Antonio,USA,2023:1-2
[171]WANG X,GUO X F.A hierarchically reconfigurable SRAM-based compute-in-memory macro for edge computing[C]//Proceedings of IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS).Hangzhou,China,2023:1-5
[172]ZHANG H,WANG M Y,MAI Y Z,et al.A high-density and reconfigurable SRAM-based digital compute-inmemory macro for low-power AI chips[J].IEEE Transactions on Circuits and Systems II:Express Briefs,2023,70(9):3589-3593
[173]SEO S.Advances and trends on on-chip compute-inmemory macros and accelerators[C]//Proceedings of the60th ACM/IEEE Design Automation Conference (DAC).San Francisco,USA,2023:1-6
[174]CHEN J,TU B,SHAO K M,et al.Auto DCIM:an automated digital CIM compiler[C]//Proceedings of the60th ACM/IEEE Design Automation Conference (DAC).San Francisco,USA,2023:1-6.
[175]WU W,CHANG C Y,WU A Y A.DE-C3:dynamic energy-aware compression for computing-In-memory-based convolutional neural network acceleration[C]//Proceedings of IEEE 36th International System-on-Chip Conference (SOCC).Santa Clara,USA,2023:1-6
[176]JOSEPH B,REDDY C,KAVITHA R K.Energy efficient memory decoder for SRAM based AI accelerator[C]//Proceedings of the 2nd International Conference on Paradigm Shifts in Communications Embedded Systems,Machine Learning and Signal Processing (PCEMS).Nagpur,India,2023:1-4
[177]ZHANG R,WANG B,CHEN J W,et al.Evaluation model for current-domain SRAM-based computing-inmemory circuits[C]//Proceedings of IEEE 16th International Symposium on Embedded Multicore/Many-Core Systems-on-Chip (MCSo C).Singapore,2023:160-165
[178]FUJIWARA H,MORI H,ZHAO C,et al.34.4 A3 nm,32.5 TOPS/W,55.0 TOPS/mm2 and 3.78 Mb/mm2 fully-digital compute-in-memory macro supporting INT12×INT12 with parallel-MAC architecture and foundry 6T-SRAM bit cell[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,USA,2024:572-574
[179]YUAN Y,YANG Y M,WANG X H,et al.34.6 A28 nm 72.12 TFLOPS/W hybrid-domain outer-product based floating-point SRAM computing-in-memory macro with logarithm bit-width residual ADC[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISS-CC).San Francisco,USA,2024:576-578
[180]WU C,SU J W,HONG L Y,et al.A floating-point 6TSRAM in-memory-compute macro using hybrid-domain structure for advanced AI edge chips[J].IEEE Journal of Solid-State Circuits,2024,59 (1):196-207
[181]CHIU C,KHWA W S,LI C Y,et al.A 22 nm 8 Mb STT-MRAM near-memory-computing macro with8 b-precision and 46.4-160.1 TOPS/W for edge-AI devices[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,CA,USA,2023:496-498
[182]LU L,MANI A,DO T.A 129.83 TOPS/W area efficient digital SOT/STT MRAM-based computing-in-memory for advanced edge AI chips[C]//Proceedings of IEEEInternational Symposium on Circuits and Systems (ISCAS).Monterey,USA,2023:1-5
[183]LI T,MA T,YOSHIKAWA K,et al.Hybrid signed convolution module with unsigned divide-and-conquer multiplier for energy-efficient STT-MRAM-based AI accelerator[J].IEEE Transactions on Very Large Scale Integration (VLSI) Systems,2023,31 (7):1078-1082.
[184]JOSHI P,RAHAMAN H.comprehensive review on Re RAM-based accelerators for deep learning[C]//Proceedings of International Symposium on Devices,Circuits and Systems (ISDCS).Higashihiroshima,Japan,2023:1-5
[185]HUANG H,WEN T H,HUNG J M,et al.A nonvolatile AI-edge processor with 4MB SLC-MLC hybridmode Re RAM compute-in-memory macro and 51.4-251 TOPS/W[C]//Proceedings of IEEE International Solid-State Circuits Conference (ISSCC).San Francisco,USA,2023:258-259.
[186]TIAN S,WANG X M,CHEN J B,et al.BIOS:a40 nm bionic sensor-defined 0.47 pJ/SOP,268.7 TSOPS/Wconfigurable spiking neuron-in-memory processor for wearable healthcare[C]//Proceedings of IEEE 49th European Solid State Circuits Conference (ESSCIRC).Lisbon,Portugal,2023:225-228
[187]XIAO H,XIE R,HU X F,et al.Brain-inspired recognition system based on multimodal in-memory computing framework for edge AI[J].IEEE Transactions on Circuits and Systems I:Regular Papers,2024,71 (5):2294-2307
[188]SHI C,SU Y B,TANG J S,et al.Counteractive coupling IGZO/CNT hybrid 2T0 DRAM accelerating RRAM-based computing-In-memory via monolithic 3 integration for edge AI[C]//Proceedings of International Electron Devices Meeting (IEDM).San Francisco,USA,2023:1-4
[189]YUAN M,CHAR Y,DAI S Q,et al.Design and demonstration of Cu/Al2O3/Cu RRAM with complementary resistance switching characteristic[C]//Proceedings of the 7th IEEE Electron Devices Technology Manufacturing Conference (EDTM).Seoul,Korea,2023:1-3
[190]GUO C,LIN W T,HOU T H,et al.FPCIM:a fully-parallel robust Re RAM CIM processor for edge AI devices[C]//Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS).Monterey,USA,2023:1-5.
[191]BURR W,NARAYANAN P,AMBROGIO S,et al.Phase change memory-based hardware accelerators for deep neural networks (invited)[C]//Proceedings of IEEESymposium on VLSI Technology and Circuits (VLSI Technology and Circuits).Kyoto,Japan,2023:1-2
[192]STURM D,MOAZENI S.Scalable coherent optical crossbar architecture using PCM for AI acceleration[C]//Proceedings of Design,Automation Test in Europe Conference&Exhibition (DATE).Antwerp,Belgium,2023:1-6
[193]YANG Y,LIU K H,DUAN Y R,et al.Three challenges in Re RAM-based process-in-memory for neural network[C]//Proceedings of IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS).Hangzhou,China,2023:1-5
[194]HSU H,WEN T H,HUANG W H,et al.A nonvolatile AI-edge processor with SLC-MLC hybrid Re RAM compute-in-memory macro using current-voltage-hybrid readout scheme[J].IEEE Journal of Solid-State Circuits,2024,59 (1):116-127
[195]KARIMZADEH F,RAYCHOWDHURY A.Twofold sparsity:joint bit-and network-level sparsity for energyefficient deep neural network using RRAM based computein-memory[J].IEEE Access,2024,12:35125-35134
[196]AN J,ZHOU Z D,WANG L F,et al.Write-verify-free MLC RRAM using nonbinary encoding for AI weight storage at the edge[J].IEEE Transactions on Very Large Scale Integration (VLSI) Systems,2024,32 (2):283-290
基本信息:
DOI:10.13250/j.cnki.wndz.25040101
中图分类号:TN40;TP18
引用信息:
[1]赵正平.人工智能大语言模型和AI芯片的新进展(续)[J].微纳电子技术,2025,62(04):7-39.DOI:10.13250/j.cnki.wndz.25040101.
基金信息: