SPEAR NeXT：面向光谱时间地球表征学习的多时间跨度因果潜变量预测
SPEAR NeXT Causal Latent Forecasting Across Multiple Horizons for Spectral Temporal Earth Representation Learning
浏览论文内容
中文总结 AI 辅助
SPEAR NeXT提出一种紧凑的逐像素多模态光谱时间基础模型，通过因果掩蔽Transformer进行仅基于过去的多时间跨度潜在地球状态预测，以捕捉地球观测的动态特性。
中文摘要 AI 辅助
地球观测本质上是动态的，然而许多基础模型中的时间信息是通过重建、不变性或回顾性序列总结来学习的。SPEAR NeXT被引入作为一个紧凑的逐像素多模态光谱时间基础模型，其中时间自监督被表述为仅基于过去的、多时间跨度潜在地球状态预测。瞬时状态首先由预训练的SPEAR模型从光学、雷达和环境观测中编码为紧凑的32维嵌入。然后，它们的时间演化由一个因果掩蔽的Transformer建模，该Transformer从先前的观测中预测多个未来的潜在状态。相对时间顺序使用旋转位置嵌入表示，而月份和年份嵌入编码季节相位和年际背景。
英文摘要
Earth observation is inherently dynamic, yet temporal information in many foundation models is learned through reconstruction, invariance, or retrospective sequence summarization. SPEAR NeXT is introduced as a compact pixel-wise multimodal spectral temporal foundation model in which temporal self supervision is formulated as past only, multi horizon latent Earth state prediction. Instantaneous states are first encoded by the pretrained SPEAR model from optical, radar, and environmental observations into compact 32 dimensional embeddings. Their temporal evolution is then modeled by a causally masked Trans former that predicts multiple future latent states from pre ceding observations. Relative temporal order is represented using Rotary Position Embeddings, while month and year embeddings encode seasonal phase and interannual con text.
发表机构
- Plaksha University（普拉克沙大学）
- Purdue University（普渡大学）
机构由 AI 辅助整理，请以论文原文为准。
