Research map / Large language models / Language, safety and interpretability
Translating Signals to Languages for sEMG-Based Activity Recognition
Ming Wang, Haoxuan Qu, Qiuhong Ke, Wei Zhou, Hossein Rahmani, Jun Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add iterated learning (decoder renewal) (ablation) | LLM-sEMG · GRABMyo overall | accuracy (%): 79.59 -> 95.14 | Improved |
| add decoder warm-up (ablation) | LLM-sEMG · GRABMyo overall | accuracy (%): 91.65 -> 95.14 | Improved |
| add all human-linguistic biases (ablation) | LLM-sEMG · GRABMyo overall | accuracy (%): 87.35 -> 95.14 | Improved |
| add Zipf prior (ablation) | LLM-sEMG · GRABMyo overall | accuracy (%): 90.65 -> 95.14 | Improved |
| add context-sensitivity constraint (ablation) | LLM-sEMG · GRABMyo overall | accuracy (%): 92.13 -> 95.14 | Improved |
| add inter-generational preserving bias (ablation) | LLM-sEMG · GRABMyo overall | accuracy (%): 93.85 -> 95.14 | Improved |
| add residual-based adaptive token allocation (ablation) | LLM-sEMG · GRABMyo overall | accuracy (%): 92.60 -> 95.14 | Improved |
| add residual-based allocation (ablation) | LLM-sEMG · GRABMyo single-finger | accuracy (%): 89.75 -> 92.25 | Improved |
| add residual-based allocation (ablation) | LLM-sEMG · GRABMyo multi-finger | accuracy (%): 91.20 -> 94.50 | Improved |
| add residual-based allocation (ablation) | LLM-sEMG · GRABMyo wrist | accuracy (%): 93.74 -> 95.19 | Improved |
| add residual-based allocation (ablation) | LLM-sEMG · GRABMyo rest | accuracy (%): 95.87 -> 96.82 | Improved |
| use LLM-sEMG instead of STET | LLM-sEMG · GRABMyo overall | accuracy (%): 90.76 -> 95.14 | Improved |
Closest papers by meaning
- emg2qwerty: A Large Dataset with Baselines for Touch Typing using Surface Electromyography (NeurIPS 2024 · similarity 0.88)
- LLMs are Good Action Recognizers (CVPR 2024 · similarity 0.85)
- SpGesture: Source-Free Domain-adaptive sEMG-based Gesture Recognition with Jaccard Attentive Spiking Neural Network (NeurIPS 2024 · similarity 0.85)
- SplashNet: Split‑and‑Share Encoders for Accurate and Efficient Typing with Surface Electromyography (NeurIPS 2025 · similarity 0.84)
- Bridging Gait Recognition and Large Language Models Sequence Modeling (CVPR 2025 · similarity 0.84)
- emg2pose: A Large and Diverse Benchmark for Surface Electromyographic Hand Pose Estimation (NeurIPS 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
