Research map / Label-efficient and robust learning / Self-supervised and contrastive
Interpretable Self-Supervised Facial Micro-Expression Learning to Predict Cognitive State and Neurological Disorders
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use self-supervised pretraining before downstream classification | CNN-A with 3FC · AWS stuttering disfluency, 10% training data | accuracy: 65.50±2.09 -> 71.75±3.92 | Helped |
| Use self-supervised pretraining before downstream classification | CNN-A with 3FC · AWS stuttering disfluency, 100% training data | accuracy: 72.76±1.38 -> 75.27±1.11 | Helped |
| Reduce the range of signal-transformation factors (ablation) | Self-supervised CNN models · AWS stuttering disfluency, across training data amounts | accuracy change: Table 3 reports ∼1.9% to ∼8.2% degradation in most settings | Hurt |
| Reduce the range of signal-transformation factors (ablation) | CNN-A RF · AWS stuttering disfluency, 50% training data | accuracy change: ∼-0.8% | Helped |
| Increase embedding dimensions from 256 to 970 (ablation) | CNN-B and CNN-C with 3FC · AWS stuttering disfluency, 100% and 50% training data | accuracy change: ∼-3.7%, ∼-0.9%, ∼-4.8%, ∼-4.5% | Hurt |
| Increase embedding dimensions from 256 to 970 (ablation) | CNN-C RF · AWS stuttering disfluency, 10% training data | accuracy change: ∼11.0% | Helped |
| Increase embedding dimensions from 256 to 970 (ablation) | CNN-B and CNN-C downstream classifiers · AWS stuttering disfluency, across training data amounts | accuracy change: Table 4 reports changes from ∼-4.8% to ∼11.0% | Mixed |
Closest papers by meaning
- Self-Supervised Learning of Brain Dynamics from Broad Neuroimaging Data (NeurIPS 2022 · similarity 0.87)
- Understanding Unique Behavioral Patterns through Multimodal Analysis of Eye-Hand Coordination in Autistic Children (Student Abstract) (AAAI 2025 · similarity 0.85)
- Exploring Behavior-Relevant and Disentangled Neural Dynamics with Generative Diffusion Models (NeurIPS 2024 · similarity 0.84)
- BRAID: Input-driven Nonlinear Dynamical Modeling of Neural-Behavioral Data (ICLR 2025 · similarity 0.83)
- SPINT: Spatial Permutation-Invariant Neural Transformer for Consistent Intracortical Motor Decoding (NeurIPS 2025 · similarity 0.83)
- Electromyography-Informed Facial Expression Reconstruction for Physiological-Based Synthesis and Analysis (CVPR 2025 · 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.