arXiv：cs.LG· Tianhao Wu, Jiawei Yu, Xiangdong Hu, Xu Wu, Hongyu Wu, Sidharth Anupkrishnan, Amirmohammad Radmehr, Yi Chen Liu, Phuc Nguyen, Yi Wu, Jian Liu·· 4 小时前
EMG-FM-Bench：面向肌电信号的基础模型迁移与适配综合基准
EMG-FM-Bench: A Comprehensive Benchmark for Foundation Model Transfer and Adaptation on Electromyography
AI 导读
EMG-FM-Bench 是面向肌电信号（EMG）的基础模型迁移与适配基准，统一了 20 个公开数据集、超 100 万个 EMG 片段，评测九种预训练基础模型。基准发现预训练收益因模型而异并非普遍存在，新用户上性能下降，五样本适配可在 70.2% 的模型-数据集组合上提升 macro-F1，但只能部分恢复损失的性能。
正文
Authors:Tianhao Wu, Jiawei Yu, Xiangdong Hu, Xu Wu, Hongyu Wu, Sidharth Anupkrishnan, Amirmohammad Radmehr, Yi Chen Liu, Phuc Nguyen, Yi Wu, Jian Liu
Abstract:Foundation models (FMs) are increasingly being developed for general time series and physiological signals, yet their transferability to downstream physiological tasks remains poorly understood. This question is particularly challenging for electromyography (EMG), where signal distributions vary substantially across users, sensing configurations, acquisition hardware, and downstream tasks. We introduce EMG-FM-Bench, a systematic benchmark for studying foundation-model transfer and adaptation on EMG. EMG-FM-Bench unifies 20 public datasets with over 1 million EMG segments and evaluates nine pretrained foundation models across four questions: how pretrained models perform when frozen or fully fine-tuned, how much pretraining helps compared with training the same model from scratch, how well models generalize to new users with limited labeled data, and how performance changes across different EMG tasks. Across the benchmark, linear probing provides useful information about pretrained representations, but full fine-tuning can substantially change downstream EMG performance. Comparing each pretrained model with the same model trained from scratch shows that the benefit of pretraining varies substantially across models and is not universal. Performance decreases when models are evaluated on new users, while five-shot adaptation improves macro-F1 in 70.2% of evaluated model-dataset combinations but recovers only part of the lost performance. Model performance is highly consistent between upper- and lower-limb classification and remains strongly correlated with continuous EMG-to-text decoding. Together, these results provide a systematic view of when pretrained time-series models transfer effectively to EMG and how their performance depends on fine-tuning, user variation, and downstream task.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.06450 [cs.LG] |
| (or arXiv:2610.06450v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.06450 arXiv-issued DOI via DataCite |
Submission history
From: Tianhao Wu [view email]
[v1]
Mon, 5 Oct 2026 14:53:56 UTC (155 KB)
[v2]
Tue, 13 Oct 2026 18:42:09 UTC (151 KB)
来源：arXiv：cs.LG · arxiv.org