M3-Former：融合混合专家的多模态长期船舶轨迹预测
原标题：M3-Former: Multimodal Transformer with Mixture-of-Experts for Long-Term Vessel Trajectory Prediction
AI 摘要
该论文提出 M3-Former，一个结合大语言模型与多模态 Transformer 的船舶轨迹预测框架，用于长期（1-4 小时）轨迹预测。方法将船舶静态属性与航行意图作为语义先验，通过预训练 LLM 编码并与动态轨迹特征对齐，同时引入双粒度混合专家（MoE）架构分别建模全局航线规划与局部机动行为，并设计转向加权交叉熵损失缓解稀疏转向样本的长尾问题。在丹麦真实 AIS 数据集上，4 小时预测任务中 ADE 和 FDE 分别较最强基线降低 4.4% 和 5.1%，代码已开源。
正文节选
M3-Former: Multimodal Transformer with Mixture-of-Experts for Long-Term Vessel Trajectory Prediction Abstract To address the challenges of behavioral multimodality, limited semantic utilization, and long-term error accumulation in vessel trajectory prediction, this paper proposes M3-Former, a multimodal trajectory prediction framework enhanced by large language models (LLMs). The proposed framework incorporates vessel static attributes and navigational intent as semantic priors for long-term tra