{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,4]],"date-time":"2026-03-04T21:07:24Z","timestamp":1772658444290,"version":"3.50.1"},"reference-count":16,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,2,15]],"date-time":"2026-02-15T00:00:00Z","timestamp":1771113600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,2,15]],"date-time":"2026-02-15T00:00:00Z","timestamp":1771113600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,2,15]]},"DOI":"10.1109\/isscc49663.2026.11409285","type":"proceedings-article","created":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T20:50:24Z","timestamp":1772571024000},"page":"312-314","source":"Crossref","is-referenced-by-count":0,"title":["SMoLPU: 122.1\u00b5J\/Token Sparse MoE-Based Speculative Decoding Language Processing Unit with Adaptive-Offload NPU-CIM Core"],"prefix":"10.1109","author":[{"given":"Sangwoo","family":"Ha","sequence":"first","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jingu","family":"Lee","sequence":"additional","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Youngjin","family":"Moon","sequence":"additional","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sunjoo","family":"Whang","sequence":"additional","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wooyoung","family":"Jo","sequence":"additional","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gwangtae","family":"Park","sequence":"additional","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sangjin","family":"Kim","sequence":"additional","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Soyeon","family":"Um","sequence":"additional","affiliation":[{"name":"Massachusetts Institute of Technology,Cambridge,MA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Junha","family":"Ryu","sequence":"additional","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yurim","family":"Jo","sequence":"additional","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hoi-Jun","family":"Yoo","sequence":"additional","affiliation":[{"name":"KAIST,Daejeon,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Self-Consistency Improves Chain of Thought Reasoning in Language Models","author":"Wang","year":"2022","journal-title":"arXiv preprint"},{"key":"ref2","first-page":"421","article-title":"Slim-Llama: A 4.69 mW Large-Language-Model Processor with Binary\/Ternary Weights for Billion-Parameter Llama Model","author":"Kim","year":"2025","journal-title":"ISSCC"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/isscc49661.2025.10904658"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/isscc49661.2025.10904774"},{"key":"ref5","first-page":"368","article-title":"C-Transformer: A $2.6-18.1 \\mu \\mathrm{J} \/$ Token Homogeneous DNN-Transformer\/Spiking-Transformer Processor with Big-Little Network and Implicit Weight Generation for Large Language Models","author":"Kim","year":"2024","journal-title":"ISSCC"},{"key":"ref6","article-title":"EAGLE-3: Scaling Up Inference Acceleration of Large Language Models via Training-Time Test","author":"Li","year":"2025","journal-title":"arXiv preprint"},{"key":"ref7","article-title":"DeepSeek-V2: A Strong, Economical, and Efficient Mixture-Of-Experts Language Model","author":"Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref8","article-title":"OLMoE: Open Mixture-Of-Experts Language Models","author":"Muennighoff","year":"2024","journal-title":"arXiv preprint"},{"key":"ref9","article-title":"Qwen3 Technical Report","author":"Yang","year":"2025","journal-title":"arXiv preprint"},{"key":"ref10","first-page":"87","article-title":"AWQ: Activation-Aware Weight Quantization for On-Device LLM Compression and Acceleration","author":"Lin","year":"2024","journal-title":"MLSys"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/isscc42615.2023.10067779"},{"key":"ref12","doi-asserted-by":"crossref","DOI":"10.23919\/VLSITechnologyandCir65189.2025.11075101","article-title":"CELLA: A 28nm Compute-Memory Co-Optimized Real-Time Digital CIM-Based Edge LLM Accelerator with 1.78 ms -Response in Prefill and 31.32 Token\/s in Decoding","volume-title":"IEEE Symp. VLSI Circuits","author":"Wu","year":"2025"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/jssc.2024.3402174"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.23919\/vlsitechnologyandcir65189.2025.11074939"},{"key":"ref15","first-page":"578","article-title":"A 28 nm 2.4 Mb\/mm2 6.9-16.3 TOPS\/mm 2 eDRAM-LUT-Based Digital-Computing-in-Memory Macro with In-Memory Encoding and Refreshing","author":"He","year":"2024","journal-title":"ISSCC"},{"key":"ref16","first-page":"46595","article-title":"Judging LLM-as-a-Judge with MT-bench and Chatbot Arena","author":"Zheng","year":"2023","journal-title":"NeurIPS"}],"event":{"name":"2026 IEEE International Solid-State Circuits Conference (ISSCC)","location":"San Francisco, CA, USA","start":{"date-parts":[[2026,2,15]]},"end":{"date-parts":[[2026,2,19]]}},"container-title":["2026 IEEE International Solid-State Circuits Conference (ISSCC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11408863\/11408946\/11409285.pdf?arnumber=11409285","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,4]],"date-time":"2026-03-04T20:47:36Z","timestamp":1772657256000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11409285\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,15]]},"references-count":16,"URL":"https:\/\/doi.org\/10.1109\/isscc49663.2026.11409285","relation":{},"subject":[],"published":{"date-parts":[[2026,2,15]]}}}