---
title: "GEM-3：用于全球天气预报的时间步长条件 Transformer"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.06241v1"
permalink: "https://feed.astralor.com/items/itm_88853788cbe5a19dd528f4eb"
published: "2026-08-06T16:27:54.000Z"
score: 66
channel: "ai"
tags: ["GEM-3","Transformer","天气预报","时间步长条件","邻域注意力"]
---

# GEM-3：用于全球天气预报的时间步长条件 Transformer

## AI 点评

> 这篇论文对气象 AI 领域很有启发，用一套权重灵活切换时间步长，直接回应了业务中短临与中长期预报的矛盾。对做预报模型研发的团队尤其重要，值得跟踪后续版本与真实业务验证。

## 速览

现有机器学习天气预报模型依赖固定的自回归时间步长，步长选择面临精细分辨率与误差累积的权衡。GEM-3 通过显式多时间步长推理，用一组权重在推理时灵活配置步长，兼顾短临预报可用性与中长期稳定滚动。混合步长训练提升 rollout 稳定性，模型仅约 1.34 亿参数，在中等范围概率技能、扩展滚动、训练推理效率和决策相关诊断方面兼顾实用。

## 关键点

- GEM-3 采用一组权重即可在推理时配置模型时间步长，从 1 小时到 24 小时灵活切换。
- 混合时间步长训练相比专用步长模型持续提升 rollout 稳定性。
- 基于邻域注意力 Transformer，参数量约 1.34 亿，轻量高效。
- 实现近 SOTA 的中期概率预报技能，同时支持扩展范围稳定滚动和决策相关诊断。

## 全文

Existing machine-learning weather forecasting models rely on predetermined and fixed autoregressive timesteps. The choice of model timestep involves a fundamental trade-off: shorter timesteps (e.g. 1 to 6 hours) finely resolve atmospheric dynamics within the diurnal cycle but increase error accumulation for a given forecast horizon, while longer timesteps (e.g. 24 hours) reduce error accumulation but limit the usability of short-range forecasts where sub-daily predictability is high. In this work, we present GEM-3, a probabilistic global weather model that addresses this trade-off through explicit multi-timestep inference. With a single set of trained weights, the model timestep can be configured at inference time to balance predictability and usability across a broad forecast horizon. Additionally, we find that mixed-timestep training consistently improves rollout stability relative to timestep-specialist models. Under the hood, GEM-3 is a lightweight neighborhood-attention transformer with ~134M parameters on an equirectangular grid with a number of architectural advancements beyond its predecessor GEM-2. The result is a practical forecasting system that couples near-SOTA medium-range probabilistic skill, stable extended-range rollouts, efficient training and inference, and decision-relevant diagnostics.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_88853788cbe5a19dd528f4eb