{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T06:10:27Z","timestamp":1773814227079,"version":"3.50.1"},"reference-count":40,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"4","license":[{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"NSF","award":["CCF-2210754"],"award-info":[{"award-number":["CCF-2210754"]}]},{"name":"NSF","award":["CCF-2312508"],"award-info":[{"award-number":["CCF-2312508"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Comput."],"published-print":{"date-parts":[[2026,4]]},"DOI":"10.1109\/tc.2026.3666458","type":"journal-article","created":{"date-parts":[[2026,2,20]],"date-time":"2026-02-20T21:16:29Z","timestamp":1771622189000},"page":"1349-1362","source":"Crossref","is-referenced-by-count":0,"title":["TRACE: Unlocking Effective CXL Bandwidth via Lossless Compression and Precision Scaling"],"prefix":"10.1109","volume":"75","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3177-5071","authenticated-orcid":false,"given":"Rui","family":"Xie","sequence":"first","affiliation":[{"name":"Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-7975-0102","authenticated-orcid":false,"given":"Asad Ul","family":"Haq","sequence":"additional","affiliation":[{"name":"Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-4718-8825","authenticated-orcid":false,"given":"Yunhua","family":"Fang","sequence":"additional","affiliation":[{"name":"Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-8535-7911","authenticated-orcid":false,"given":"Linsen","family":"Ma","sequence":"additional","affiliation":[{"name":"Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-7058-9506","authenticated-orcid":false,"given":"Zirak Burzin","family":"Engineer","sequence":"additional","affiliation":[{"name":"Wiseburn Da Vinci Science, El Segundo, CA, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0792-8146","authenticated-orcid":false,"given":"Liu","family":"Liu","sequence":"additional","affiliation":[{"name":"Rensselaer Polytechnic Institute, Troy, NY, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-8005-0043","authenticated-orcid":false,"given":"Tong","family":"Zhang","sequence":"additional","affiliation":[{"name":"Rensselaer Polytechnic Institute, Troy, NY, USA"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/3669900"},{"key":"ref2","article-title":"CXL 101: Opportunities and challenges for compute express link","author":"Consortium","year":"2024"},{"key":"ref3","first-page":"37","article-title":"Managing memory tiers with {CXL} in virtualized environments","volume-title":"Proc. 18th USENIX Symp. Oper. Syst. Des. Implementation (OSDI)","author":"Zhong","year":"2024"},{"key":"ref4","article-title":"NVIDIA h200 tensor core GPU.\u201d"},{"key":"ref5","article-title":"The memory wall: Past, present, and future of dram","author":"Patel","year":"2024"},{"key":"ref6","article-title":"PCI express 7.0 specification, version 0.5.\u201d"},{"key":"ref7","article-title":"JEDEC Releases New LPDDR6 Standard to Enhance Mobile and AI Memory Performance","year":"2024"},{"key":"ref8","article-title":"GPTQ: Accurate post-training quantization for generative pre-trained transformers","author":"Frantar","year":"2022"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3714983.3714987"},{"key":"ref10","article-title":"Mixture-of-depths: Dynamically allocating compute in transformer-based language models","author":"Raposo","year":"2024"},{"key":"ref11","first-page":"22\u2009137","article-title":"Deja Vu: Contextual sparsity for efficient LLMs at inference time","volume-title":"Proc. PMLR","author":"Liu","year":"2023"},{"key":"ref12","article-title":"Quest: Query-aware sparsity for efficient long-context LLM inference","author":"Tang","year":"2024"},{"key":"ref13","article-title":"Kivi: A tuning-free asymmetric 2bit quantization for KV cache","author":"Liu","year":"2024"},{"key":"ref15","article-title":"Zstandard (ZSTD). Facebook"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6239"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3474381"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P16-1144"},{"key":"ref19","article-title":"Measuring massive multitask language understanding","author":"Hendrycks","year":"2020"},{"key":"ref20","article-title":"NVIDIA GB200 NVL72.\u201d"},{"key":"ref21","article-title":"NVIDIA DGX B200.\u201d"},{"key":"ref22","article-title":"Gpt-Oss-120b & Gpt-Oss-20b model card","author":"Agarwal","year":"2025"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.findings-emnlp.488"},{"key":"ref25","article-title":"The Llama 3 herd of models","author":"Dubey","year":"2024"},{"key":"ref26","article-title":"Mixtral of experts","author":"Jiang","year":"2024"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.890"},{"key":"ref28","article-title":"OPT: Open pre-trained transformer language models","author":"Zhang","year":"2022"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/LCA.2020.2973991"},{"key":"ref30","article-title":"LoRA: Low-rank adaptation of large language models","author":"Hu","year":"2021"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.183"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1016\/j.mejo.2016.04.006"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/2808233"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1145\/2882903.2903741"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3695794.3695805"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE.2015.7113373"},{"key":"ref38","first-page":"87","article-title":"AWQ: Activation-aware weight quantization for on-device LLM compression and acceleration","volume-title":"Proc. Mach. Learn. Syst.","volume":"6","author":"Lin","year":"2024"},{"key":"ref39","first-page":"10323","article-title":"SparseGPT: Massive language models can be accurately pruned in one-shot","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Frantar","year":"2023"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001172"},{"key":"ref41","article-title":"Any-precision LLM: Low-cost deployment of multiple, different-sized LLMs","author":"Park","year":"2024"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/LCA.2024.3452699"},{"key":"ref43","article-title":"Mamba: Linear-time sequence modeling with selective state spaces","volume-title":"Proc. 1st Conf. Lang. Model.","author":"Gu","year":"2024"}],"container-title":["IEEE Transactions on Computers"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/12\/11435995\/11402914.pdf?arnumber=11402914","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T05:16:37Z","timestamp":1773810997000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11402914\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4]]},"references-count":40,"journal-issue":{"issue":"4"},"URL":"https:\/\/doi.org\/10.1109\/tc.2026.3666458","relation":{},"ISSN":["0018-9340","1557-9956","2326-3814"],"issn-type":[{"value":"0018-9340","type":"print"},{"value":"1557-9956","type":"electronic"},{"value":"2326-3814","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4]]}}}