面向人体活动识别的轻量级SensorLLM的重力感知层次路由
Gravity-Aware Hierarchical Routing for Lightweight SensorLLM on Human Activity Recognition
AI总结:
针对轻量级SensorLLM在静态活动识别上的退化问题,提出一种基于重力感知层次路由的轻量级后对齐适配方法,通过统计线索和软路由显著提升静态类别的宏F1分数。
摘要:
最近关于传感器-语言对齐的研究表明,两阶段框架可以提高可穿戴传感器人体活动识别(HAR)的语义建模能力,其中SensorLLM风格的方法首先进行运动到语言的对齐,然后微调模型用于下游任务。然而,我们的实验揭示了一个一致的失败模式:当第二阶段的主干被压缩到紧凑模型(如TinyLlama)时,动态活动的识别仍然相对较强,而低运动静态类别(如站立、坐着和躺着)的区分能力显著下降。为了解决这个问题,我们提出了一种重力感知层次路由头,作为一种轻量级的后对齐适配方法,构建在已经对齐的模型之上,而不是一个新的大规模预训练框架。该方法使用来自Chronos分词器状态的每通道均值和标准差来提取与姿势和重力方向相关的统计线索,并通过软路由自适应地结合静态专家和全专家,同时使用负载平衡损失进行稳定训练。在MHealth数据集上,该设计以最小的参数开销显著提高了宏F1分数,并且增益主要集中在静态类别上,同时保持了对动态活动的强性能。作为arXiv上的首次披露,本文仅报告了单个数据集上的结果,旨在突出核心方法,并为未来工作中的更广泛评估奠定基础。
英文摘要:
Recent studies on sensor-language alignment have shown that two-stage frameworks can improve the semantic modeling ability of wearable-sensor human activity recognition (HAR), where SensorLLM-style methods first perform motion-to-language alignment and then fine-tune the model for downstream tasks. However, our experiments reveal a consistent failure mode when the Stage 2 backbone is compressed to a compact model such as TinyLlama: recognition of dynamic activities remains relatively strong, while the discrimination of low-motion static classes such as standing, sitting, and lying degrades substantially. To address this issue, we propose a gravity-aware hierarchical routing head as a lightweight post-alignment adaptation built on top of an already aligned model, rather than a new large-scale pretraining framework. The method uses the per-channel mean and std from the Chronos tokenizer state to extract statistical cues related to posture and gravity direction, and adaptively combines a static expert and a full expert through soft routing, together with a load-balancing loss for stable training. On the MHealth dataset, this design significantly improves macro-F1 with minimal parameter overhead, and the gains are concentrated mainly on static classes while preserving strong performance on dynamic activities. As a first arXiv disclosure, the current paper reports results on a single dataset only, with the goal of highlighting the core method and laying the groundwork for broader evaluation in future work.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。