{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T05:44:54Z","timestamp":1782798294269,"version":"3.54.5"},"reference-count":31,"publisher":"IEEE","license":[{"start":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T00:00:00Z","timestamp":1779062400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T00:00:00Z","timestamp":1779062400000},"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":[[2026,5,18]]},"DOI":"10.1109\/infocom59046.2026.11571335","type":"proceedings-article","created":{"date-parts":[[2026,6,29]],"date-time":"2026-06-29T19:38:15Z","timestamp":1782761895000},"page":"1-10","source":"Crossref","is-referenced-by-count":0,"title":["GoodSpeed: Optimizing Fair Goodput with Adaptive Speculative Decoding in Distributed Edge Inference"],"prefix":"10.1109","author":[{"given":"Phuong","family":"Tran","sequence":"first","affiliation":[{"name":"The University of Sydney,School of Computer Science,Darlington,NSW,Australia,2006"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tzu-Hao","family":"Liu","sequence":"additional","affiliation":[{"name":"The University of Sydney,School of Computer Science,Darlington,NSW,Australia,2006"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Long Tan","family":"Le","sequence":"additional","affiliation":[{"name":"The University of Sydney,School of Computer Science,Darlington,NSW,Australia,2006"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tung-Anh","family":"Nguyen","sequence":"additional","affiliation":[{"name":"The University of Sydney,School of Computer Science,Darlington,NSW,Australia,2006"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Van Quan","family":"La","sequence":"additional","affiliation":[{"name":"The University of Sydney,School of Computer Science,Darlington,NSW,Australia,2006"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eason","family":"Yu","sequence":"additional","affiliation":[{"name":"The University of Sydney,School of Computer Science,Darlington,NSW,Australia,2006"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Han","family":"Shu","sequence":"additional","affiliation":[{"name":"The University of Sydney,School of Computer Science,Darlington,NSW,Australia,2006"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Choong Seon","family":"Hong","sequence":"additional","affiliation":[{"name":"Kyung Hee University,Department of Computer Science and Engineering,Yongin-si,Korea,17104"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nguyen H.","family":"Tran","sequence":"additional","affiliation":[{"name":"The University of Sydney,School of Computer Science,Darlington,NSW,Australia,2006"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Distserve: disaggregating prefill and decoding for goodput-optimized large language model serving","volume-title":"Proceedings of the 18th USENIX Conference on Operating Systems Design and Implementation, ser. OSDI\u201924","author":"Zhong"},{"key":"ref2","article-title":"Taming throughput-latency tradeoff in llm inference with sarathi-serve","volume-title":"Proceedings of the 18th USENIX Conference on Operating Systems Design and Implementation, ser. OSDI\u201924","author":"Agrawal"},{"key":"ref3","article-title":"Revisiting slo and goodput metrics in llm serving","author":"Wang","year":"2024"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref5","article-title":"The new llm bottleneck: A systems perspective on latent attention and mixture-of-experts","author":"Yun","year":"2025"},{"key":"ref6","article-title":"Fast inference from transformers via speculative decoding","volume-title":"Proceedings of the 40th International Conference on Machine Learning, ser. ICML\u201923","author":"Leviathan"},{"key":"ref7","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2024.findings-acl.456","article-title":"Unlocking efficiency in large language model inference: A comprehensive survey of speculative decoding","author":"Xia","year":"2024"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651335"},{"key":"ref9","article-title":"Break the sequential dependency of llm inference using lookahead decoding","volume-title":"Proceedings of the 41st International Conference on Machine Learning, ser. ICML\u201924","author":"Fu"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.607"},{"key":"ref11","article-title":"SWIFT: On-the-fly self-speculative decoding for LLM inference acceleration","volume-title":"The Thirteenth International Conference on Learning Representations","author":"Xia"},{"key":"ref12","article-title":"Orca: A distributed serving system for transformer-based generative models","volume-title":"USENIX Symposium on Operating Systems Design and Implementation","author":"Yu"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref14","article-title":"Deepspeed-fastgen: High-throughput text generation for llms via mii and deepspeed-inference","author":"Holmes","year":"2024"},{"key":"ref15","article-title":"Inference without interference: Disaggregate llm inference for mixed downstream workloads","author":"Hu","year":"2024"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA59077.2024.00019"},{"key":"ref17","article-title":"Hetegen: Efficient heterogeneous parallel inference for large language models on resource-constrained devices","author":"Zhao","year":"2024","journal-title":"MLSys"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3669940.3707215"},{"key":"ref19","article-title":"OPTQ: Accurate quantization for generative pre-trained transformers","volume-title":"The Eleventh International Conference on Learning Representations","author":"Frantar"},{"key":"ref20","article-title":"Awq: Activation-aware weight quantization for llm compression and acceleration","author":"Lin","year":"2024","journal-title":"MLSys"},{"key":"ref21","article-title":"Medusa: Simple llm inference acceleration framework with multiple decoding heads","volume-title":"Proceedings of the 41st International Conference on Machine Learning","author":"Cai"},{"key":"ref22","article-title":"Eagle: speculative sampling requires rethinking feature uncertainty","volume-title":"Proceedings of the 41st International Conference on Machine Learning","author":"Li"},{"key":"ref23","article-title":"Ringattention with blockwise transformers for near-infinite context","volume-title":"The Twelfth International Conference on Learning Representations","author":"Liu"},{"key":"ref24","article-title":"Prompt cache: Modular attention reuse for low-latency inference","author":"Gim","year":"2024","journal-title":"MLSys"},{"key":"ref25","article-title":"Melange: Cost efficient large language model serving by exploiting gpu heterogeneity","author":"Griggs","year":"2024"},{"key":"ref26","article-title":"ServerlessLLM: Low-Latency serverless inference for large language models","volume-title":"18th USENIX Symposium on Operating Systems Design and Implementation (OSDI 24)","author":"Fu"},{"key":"ref27","doi-asserted-by":"crossref","DOI":"10.1145\/3769102.3770608","article-title":"SLED: A Speculative LLM Decoding Framework for Efficient Edge Serving","author":"Li","year":"2025"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM55648.2025.11044522"},{"key":"ref29","article-title":"DSSD: Efficient edge-device deployment and collaborative inference via distributed split speculative decoding","volume-title":"ICML 2025 Workshop on Machine Learning for Wireless Communication and Networks (ML4Wireless)","author":"NING"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1287\/opre.1040.0156"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/3767295.3769315"}],"event":{"name":"IEEE INFOCOM 2026 - IEEE Conference on Computer Communications","location":"Tokyo, Japan","start":{"date-parts":[[2026,5,18]]},"end":{"date-parts":[[2026,5,21]]}},"container-title":["IEEE INFOCOM 2026 - IEEE Conference on Computer Communications"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11571071\/11571169\/11571335.pdf?arnumber=11571335","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T05:16:54Z","timestamp":1782796614000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11571335\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,18]]},"references-count":31,"URL":"https:\/\/doi.org\/10.1109\/infocom59046.2026.11571335","relation":{},"subject":[],"published":{"date-parts":[[2026,5,18]]}}}