arXiv：cs.LG· Zhao Meng, Yinan Cai, Siru Zhong, Juepeng Zheng, Haohuan Fu·· 4 小时前AI 评分33
GeoPrior-Mamba：用语言模型构建结构化过程先验的 Mamba 框架实现精细分辨率 XCO2 重建
GeoPrior-Mamba: Structured Process Priors with Mamba for Fine-Resolution XCO2 Reconstruction
AI 导读
GeoPrior-Mamba 通过离线语言模型构建生物圈吸收、生态系统呼吸与人为排放的确定性先验表，并通过轻量知识适配器注入多方向 Mamba 重建主干。在 2018-2020 年 OCO-2 观测上，该模型 RMSE 为 0.81 ppm、R2 为 0.93，较 CAMS 背景插值降低 RMSE 48.2%，较 Trans-XCO2 降低 3.1%。
正文
Abstract:Reconstructing fine-resolution column-averaged dry-air CO2 (XCO2) fields from sparse satellite observations requires models to infer spatial structure that is only weakly constrained by direct measurements. Existing learning-based methods typically treat environmental covariates as ordinary numerical inputs and must therefore learn heterogeneous source-sink relationships largely from sparse supervision. We introduce GeoPrior-Mamba, a multi-directional Mamba framework augmented with offline language-model-induced structured process priors. Rather than using a language model to predict XCO2, we use it before training to organize relative process knowledge for biospheric uptake, ecosystem respiration, and anthropogenic emissions into deterministic prior tables. These priors are spatially instantiated using geographic, ecological, emission-related, and seasonal information and are adaptively injected into the reconstruction backbone through a lightweight knowledge adapter. Using OCO-2 observations from 2018-2020, GeoPrior-Mamba achieves an RMSE of 0.81 ppm and an R2 of 0.93 on held-out observations, reducing RMSE by 48.2% relative to CAMS background interpolation and by 3.1% relative to Trans-XCO2 under the same evaluation protocol. Ablation experiments show a measurable contribution from the knowledge-prior branch and substantially faster convergence than the knowledge-free Mamba backbone. Independent TCCON evaluation further supports the consistency of the reconstructed fields with ground-based column CO2 measurements. These results suggest that language models can provide a practical mechanism for constructing structured process priors when globally consistent process-response representations are difficult to obtain directly, while remaining outside the numerical prediction loop.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.09456 [cs.LG] |
| (or arXiv:2610.09456v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.09456 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhao Meng [view email]
[v1]
Wed, 7 Oct 2026 05:11:52 UTC (4,045 KB)
来源：arXiv：cs.LG · arxiv.org