TSFM 碳足迹研究：Chronos-2 预训练成本数月内可摊平
Sustainability of TSFM Deployment and Pretraining: From Carbon Footprint Analysis to Mitigation via Pruning
算了一笔很实在的账：Chronos-2 什么时候比传统模型划算、剪掉 7.9% 参数还能省 9% 碳排放，做时序预测部署的可以看看。
arXiv 论文对比了 Time Series Foundation Models 与需周期性重训练的监督模型在全部署生命周期中的累计碳排放。以 Chronos-2 为例，在查询频繁或规模较大的部署中，预训练成本可在数月内摊平；而小规模或低频查询场景下，传统模型始终更便宜。研究还提出一种无梯度剪枝方法，移除单个 encoder block（占参数 7.9%）即可将 FLOPs、CO2 排放和延迟降低 6 至 9%，且在 25 个任务基准上几乎不影响 zero-shot 预测性能。
Sustainability of TSFM Deployment and Pretraining: From Carbon Footprint Analysis to Mitigation via Pruning
Time Series Foundation Models (TSFMs) offer strong zero-shot forecasting capabilities, but their high pretraining cost raises a practical sustainability question: does this upfront investment pay off over a full deployment lifecycle, compared to supervised standalone models that must be periodically retrained? We model the cumulative carbon cost of different approaches across a realistic industrial scenario: Chronos-2 amortizes its pretraining cost within months for large or frequently queried deployments, but a classical model remains cheaper indefinitely for small or rarely queried ones. For cases where reducing the TSFM's own cost matters, we show that removing a single encoder block (7.9% of parameters) via a gradient-free pruning criterion yields a consistent 6 to 9% reduction in FLOPs, CO2 emissions, and latency, at essentially no cost to zero-shot forecasting performance across a 25-task benchmark.