Research map / Time series / Causal and generative time series
MambaSL: Exploring Single-Layer Mamba for Time Series Classification
Yoo-Min Jung, Leekyung Kim
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 H1 (scaled input projection kernel k=0.02L) (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 80.215 -> 79.802 (30 datasets); 77.413 -> 78.779 (10 datasets) | Mixed |
| add H2 (modular time variance/invariance) (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 77.075 -> 79.802 | Improved |
| add H3 (remove skip connection D) (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 79.163 -> 79.802 | Improved |
| add H4 adaptive pooling instead of fully connected (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 77.722 -> 79.802 | Improved |
| add H4 adaptive pooling instead of average pooling (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 78.610 -> 79.802 | Improved |
| add H4 adaptive pooling instead of max pooling (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 78.535 -> 79.802 | Improved |
| add H4 adaptive pooling instead of last-step readout (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 76.848 -> 79.802 | Improved |
| add H1, H3, H4 on top of H2-only variant (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 77.935 -> 79.802 | Improved |
| add all four hypotheses H1-H4 to vanilla Mamba (ablation) | Mamba · UEA 30 datasets | avg accuracy: 74.243 -> 79.802 | Improved |
| make Delta, B, C all time-variant (vs all time-invariant) (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 77.400 -> 77.055 | Hurt |
| stack 2 Mamba blocks instead of 1 (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 79.819 -> 79.316 | Hurt |
| stack 3 Mamba blocks instead of 1 (ablation) | MambaSL · UEA 30 datasets | avg accuracy: 79.819 -> 79.279 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TimesNet — TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis (ICLR 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Block-Biased Mamba for Long-Range Sequence Processing (NeurIPS 2025 · similarity 0.87)
- MambaLRP: Explaining Selective State Space Sequence Models (NeurIPS 2024 · similarity 0.86)
- Mamba Modulation: On the Length Generalization of Mamba Models (NeurIPS 2025 · similarity 0.86)
- LongMamba: Enhancing Mamba's Long-Context Capabilities via Training-Free Receptive Field Enlargement (ICLR 2025 · similarity 0.85)
- Achilles' Heel of Mamba: Essential difficulties of the Mamba architecture demonstrated by synthetic data (NeurIPS 2025 · similarity 0.85)
- Mamba Can Learn Low-Dimensional Targets In-Context via Test-Time Feature Learning (NeurIPS 2026 · similarity 0.85)
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.
