遥感场景分类是地球观测和地理空间分析中的基础任务。现有方法主要遵循三种范式：特定任务的视觉分类、视觉-语言相似度匹配以及自回归多模态生成。然而，视觉分类器依赖预定义的标签空间，基于 CLIP 的方法通过静态图文对齐进行识别，而多模态大语言模型（MLLMs）在具有明确候选类别的分类任务中引入了不必要的 token 级生成。为解决这些局限性，我们提出了 RSJEV，一种用于遥感场景分类的单次前向多模态决策框架。与将分类表述为自回归文本生成的传统 MLLMs 不同，RSJEV 将场景分类重构为候选条件化的多模态判别决策过程，其中联合建模了视觉表征、任务指令和候选类别语义。具体而言，我们引入了 OnePass Decider，它提取多模态决策状态并在候选类别空间内直接估计类别概率，从而消除了自回归解码同时保留了视觉-语言交互。在三个广泛使用的遥感场景分类基准数据集（包括 UC Merced、AID 和 NWPU-RESISC45）上的大量实验表明，与代表性的基于 CNN、Transformer、Mamba、CLIP 和 MLLM 的方法相比，RSJEV 实现了更优的分类性能。此外，仅使用一个紧凑的 0.8B 参数模型，RSJEV 显著降低了推理成本，并实现了更好的精度-效率权衡。这些结果证明了状态条件化多模态决策在高效遥感图像理解中的有效性。代码将在 https://github.com/Dongtcs/RSJEV 公开。

Remote sensing scene classification is a fundamental task in Earth observation and geospatial analysis. Existing approaches mainly follow three paradigms: task-specific visual classification, vision-language similarity matching, and autoregressive multimodal generation. However, visual classifiers rely on predefined label spaces, CLIP-based methods perform recognition through static image-text alignment, and multimodal large language models (MLLMs) introduce unnecessary token-level generation for classification tasks with explicit candidate categories. To address these limitations, we propose RSJEV, a one-pass multimodal decision framework for remote sensing scene classification. Unlike conventional MLLMs that formulate classification as autoregressive text generation, RSJEV reformulates scene classification as a candidate-conditioned multimodal discriminative decision process, where visual representations, task instructions, and candidate category semantics are jointly modeled. Specifically, we introduce a OnePass Decider that extracts multimodal decision states and directly estimates category probabilities within the candidate category space, eliminating autoregressive decoding while preserving vision-language interactions. Extensive experiments on three widely used remote sensing scene classification benchmarks, including UC Merced, AID, and NWPU-RESISC45, demonstrate that RSJEV achieves superior classification performance compared with representative CNN-, Transformer-, Mamba-, CLIP-, and MLLM-based methods. Moreover, RSJEV significantly reduces inference costs and achieves a better accuracy-efficiency trade-off with only a compact 0.8B-parameter model. These results demonstrate the effectiveness of state-conditioned multimodal decision making for efficient remote sensing image understanding. The code will be available at https://github.com/Dongtcs/RSJEV.