GreenPassport：跨境AI推理的请求级碳核算
GreenPassport: Request-Level Carbon Accounting for Cross-Border AI Inference
浏览论文内容
中文总结 AI 辅助
GreenPassport提出请求级碳核算方法，关联服务、路由等输入，估算碳排放并选择报告级别，在测试中误差显著降低，并展示清洁电力场景的减排优势。
中文摘要 AI 辅助
AI推理常常跨越区域边界，因为提示词被发送到远程数据中心，而生成的令牌返回给用户。区域平均值无法代表由此产生的服务硬件、电力和网络传输差异。请求级核算需要为服务、服务站点、路由、本地比较器、不确定性和数据定义一个共同边界。绿色护照碳核算（GPCA）将这些输入与每个请求关联起来。它估算服务和路由的碳排放，然后从可用文档中选择报告级别。我们的公共数据实现涵盖了数据中心实例、加速器、模型家族、电力组合、路由和云区域碳强度。与六个核算基线和四个能源预测基线相比，在对齐的加速器能源边界下，GPCA相对于EcoLogits将中位绝对百分比误差降低了56.3%，中位绝对误差降低了15.5%。在确定性一致性测试中，它产生了零规则高估。在买方案例中，清洁电力CN-West情景产生了0.0148克二氧化碳当量/请求，比本地服务的0.1220克二氧化碳当量/请求低88%。
英文摘要
AI inference often crosses regional boundaries as prompts travel to remote data centers and generated tokens return to users. Regional averages cannot represent the resulting differences in serving hardware, electricity, and network delivery. Request-level accounting needs a common boundary for the service, serving site, route, local comparator, uncertainty, and data provenance. GreenPassport Carbon Accounting (GPCA) associates these inputs with each request. It estimates serving and route carbon, then selects a reporting level from the available documentation. Our public-data implementation covers data-center instances, accelerators, model families, electricity mixes, routes, and cloud-region carbon intensity. Against six accounting baselines and four energy-prediction baselines, GPCA reduced median absolute percentage error by 56.3% and median absolute error by 15.5% relative to EcoLogits under the aligned accelerator-energy boundary. It produced zero rule overstatement in the deterministic conformance tests. In the buyer case, the clean-electricity CN-West scenario produced 0.0148 gCO2e per request, 88% below the local service at 0.1220 gCO2e per request.
