小数据下用混合模型学躯干运动，提升跌倒风险预测
Sonata: A Hybrid World Model for Inertial Kinematics under Clinical Data Scarcity
- 用未来状态预测替代原始信号重建，适配临床小样本
- 377万参数，在14个队列中表现更优，跨队列迁移强
- 适合神经评估的可部署式可穿戴运动建模，小数据友好
我们提出Sonata，一种针对临床数据稀缺场景下的六轴躯干惯性测量单元（IMU）表示学习的紧凑潜在世界模型。临床队列通常仅含数十至数百名患者，难以支撑大规模掩码重建任务。Sonata为377万参数的混合模型，基于九个公开数据集（共739名受试者，19万个时间窗口）进行预训练，采用潜在世界模型目标预测未来状态而非重建原始传感器信号。在相同主干网络下与自回归预测基线（MAE）对比，Sonata在冻结探针测试中展现出更强的临床判别能力、前瞻性跌倒风险预测性能及跨队列迁移表现，且生成更高秩、更结构化的潜在表示。该模型仅377万参数，兼容设备端可穿戴推理，推动神经评估通用运动世界模型的发展。
原文摘要 · Abstract (English)
We introduce Sonata, a compact latent world model for six-axis trunk IMU representation learning under clinical data scarcity. Clinical cohorts typically comprise tens to hundreds of patients, making web-scale masked-reconstruction objectives poorly matched to the problem. Sonata is a 3.77 M-parameter hybrid model, pre-trained on a harmonised corpus of nine public datasets (739 subjects, 190k windows) with a latent world-model objective that predicts future state rather than reconstructing raw sensor traces. In a controlled comparison against a matched autoregressive forecasting baseline (MAE) on the same backbone, Sonata yields consistently stronger frozen-probe clinical discrimination, prospective fall-risk prediction, and cross-cohort transfer across a 14-arm evaluation suite, while producing higher-rank, more structured latent representations. At 3.77 M parameters the model is compatible with on-device wearable inference, offering a step toward general kinematic world models for neurological assessment.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。