{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,21]],"date-time":"2025-11-21T05:56:43Z","timestamp":1763704603982,"version":"3.45.0"},"reference-count":40,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,26]],"date-time":"2025-10-26T00:00:00Z","timestamp":1761436800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,26]],"date-time":"2025-10-26T00:00:00Z","timestamp":1761436800000},"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":[[2025,10,26]]},"DOI":"10.1109\/iccad66269.2025.11240889","type":"proceedings-article","created":{"date-parts":[[2025,11,20]],"date-time":"2025-11-20T18:39:34Z","timestamp":1763663974000},"page":"1-9","source":"Crossref","is-referenced-by-count":0,"title":["LP-Spec: Leveraging LPDDR PIM for Efficient LLM Mobile Speculative Inference with Architecture-Dataflow Co-Optimization"],"prefix":"10.1109","author":[{"given":"Siyuan","family":"He","sequence":"first","affiliation":[{"name":"School of Integrated Circuits Peking University,Beijing,China"}]},{"given":"Zhantong","family":"Zhu","sequence":"additional","affiliation":[{"name":"School of Integrated Circuits Peking University,Beijing,China"}]},{"given":"Yandong","family":"He","sequence":"additional","affiliation":[{"name":"School of Integrated Circuits Peking University,Beijing,China"}]},{"given":"Tianyu","family":"Jia","sequence":"additional","affiliation":[{"name":"School of Integrated Circuits Peking University,Beijing,China"}]}],"member":"263","reference":[{"article-title":"Language Models are Few-Shot Learners","year":"2020","author":"Brown","key":"ref1"},{"year":"2022","key":"ref2","article-title":"Introducing ChatGPT"},{"key":"ref3","article-title":"Github Copilot"},{"year":"2024","key":"ref4","article-title":"Apple Intelligence comes to iPhone, iPad, and Mac starting next month"},{"year":"2024","key":"ref5","article-title":"Use Live Translate on Galaxy phones and tablets"},{"article-title":"LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale","year":"2022","author":"Dettmers","key":"ref6"},{"article-title":"QLoRA: Efficient Finetuning of Quantized LLMs","year":"2023","author":"Dettmers","key":"ref7"},{"article-title":"SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models","year":"2024","author":"Xiao","key":"ref8"},{"article-title":"LLM-Pruner: On the Structural Pruning of Large Language Models","year":"2023","author":"Ma","key":"ref9"},{"article-title":"SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot","year":"2023","author":"Frantar","key":"ref10"},{"article-title":"A Simple and Effective Pruning Approach for Large Language Models","year":"2024","author":"Sun","key":"ref11"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651335"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3676536.3676695"},{"key":"ref14","first-page":"19274","article-title":"Fast Inference from Transformers via Speculative Decoding","volume-title":"40th International Conference on Machine Learning (ICML)","author":"Leviathan"},{"article-title":"Accelerating Large Language Model Decoding with Speculative Sampling","year":"2023","author":"Chen","key":"ref15"},{"article-title":"Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads","volume-title":"International Conference on Machine Learning (ICML)","author":"Cai","key":"ref16"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.257"},{"article-title":"LLM Inference Unveiled: Survey and Roofline Model Insights","year":"2024","author":"Yuan","key":"ref18"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42614.2022.9731711"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2022.3164651"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42614.2022.9731639"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42615.2023.10067271"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/HCS59251.2023.10254711"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640422"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2857044"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2020.3011265"},{"year":"2023","key":"ref27","article-title":"Low Power Double Data Rate (LPDDR) 5\/5X"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651380"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651352"},{"article-title":"Blockwise Parallel Decoding for Deep Autoregressive Models","year":"2018","author":"Stern","key":"ref30"},{"article-title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","year":"2023","author":"Touvron","key":"ref31"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00013"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/3676641.3716009"},{"key":"ref34","article-title":"Pimsimulator"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA59077.2024.00082"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1016\/j.mejo.2016.04.006"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/HOTCHIPS.2019.8875680"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42613.2021.9366050"},{"article-title":"Judging LLM-as-a-judge with MT-Bench and Chatbot Arena","year":"2023","author":"Zheng","key":"ref39"},{"article-title":"Stanford Alpaca: An Instruction-following LLaMA model","year":"2023","author":"Taori","key":"ref40"}],"event":{"name":"2025 IEEE\/ACM International Conference On Computer Aided Design (ICCAD)","start":{"date-parts":[[2025,10,26]]},"location":"Munich, Germany","end":{"date-parts":[[2025,10,30]]}},"container-title":["2025 IEEE\/ACM International Conference On Computer Aided Design (ICCAD)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11240608\/11240621\/11240889.pdf?arnumber=11240889","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,21]],"date-time":"2025-11-21T05:45:50Z","timestamp":1763703950000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11240889\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,26]]},"references-count":40,"URL":"https:\/\/doi.org\/10.1109\/iccad66269.2025.11240889","relation":{},"subject":[],"published":{"date-parts":[[2025,10,26]]}}}