arXiv cs.CL· Viju Sudhi, Giulio Billi, Andreea Scrob, Regina Stodden, Philipp Cimiano·· 12 小时前AI 评分31
Tug of Watts：大语言模型在医疗问答任务中的能耗与准确率权衡研究
Tug of Watts: A study on the Energy - Accuracy Trade-off of Large Language Models on Medical Question Answering Tasks
AI 导读
研究基于 OpenMedical-LLM 榜单，考察 3 个模型家族共 13 个 LLM（270M 至 14B 参数）在医疗问答任务中的能耗与准确率权衡，并提出仅分别报告能耗和准确率指标不足以支撑审慎的权衡评估，还需系统构建并解读联合权衡指标。
正文 · 原文
Abstract:Large Language Models (LLMs) are increasingly used in medical question answering (QA) systems. While recent works aim at incrementally improving the downstream task accuracy, an empirical estimation of the models' energy consumption and the associated carbon footprints is often overlooked. In this study, we demonstrate how practitioners can estimate energy consumption and arrive at reliable emission estimates for medical QA tasks available in the OpenMedical-LLM leaderboard by investigating 13 LLMs from 3 model families ranging from 270M to 14B parameters. We discuss how one can critically interpret energy versus accuracy trade-offs for a set of models for a given task. With this work, we underline that reporting disjoint energy and accuracy metrics alone does not suffice for careful trade-off assessments, but it also demands systematic construction and interpretation of joint trade-off metrics. We envisage such studies aiding practitioners in adopting better and more efficient models for their task and helping them become active participants in building sustainable LLM pipelines.
| Comments: | Accepted to "Workshop on GREEN-AI: Grounding Research on Energy- and resource-Efficient Networks for sustainable AI", Co-Located with CIKM 2026, November 08, 2026, Rome, Italy |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.02889 [cs.CL] |
| (or arXiv:2611.02889v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.02889 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Viju Sudhi [view email]
[v1]
Mon, 2 Nov 2026 10:23:57 UTC (1,084 KB)
来源：arXiv cs.CL · arxiv.org