Research map / Time series / Long-term forecasting
Statistical Test for Attention in Transformers for Images and Time Series
Tomohiro Shiraishi, Daiki Miwa, Teruyuki Katsuoka, Vo Nguyen Le Duy, Shuichi Nishino, Kouichi Taji, Ichiro Takeuchi
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 |
|---|---|---|---|
| use the proposed adaptive selective inference test instead of the naive test | Transformer attention-region test · Synthetic images and time series (type I error) | type I error rate | Improved |
| use the adaptive selective inference test instead of the permutation test | Transformer attention-region test · Synthetic images and time series (type I error) | type I error rate | Improved |
| use the adaptive test instead of Bonferroni correction | Transformer attention-region test · Synthetic images and time series (power, n=256, base) | statistical power | Improved |
| use adaptive grid search instead of fixed or combination grid strategies | Selective inference computation · Synthetic images and time series (computation time) | computation time | Improved |
| use selective p-values instead of naive p-values | Transformer classifier on BraTS 2023 MRI and EEG RSVP · Real-world MRI and EEG negative samples | p-value on negative samples | Improved |
Closest papers by meaning
- Statistical Test for Attention Maps in Vision Transformers (ICML 2024 · similarity 0.92)
- Hierarchical Self-Attention: Generalizing Neural Attention Mechanics to Multi-Scale Problems (NeurIPS 2025 · similarity 0.86)
- Transformer brain encoders explain human high-level visual responses (NeurIPS 2025 · similarity 0.85)
- Selective Attention Improves Transformer (ICLR 2025 · similarity 0.85)
- Linear Log-Normal Attention with Unbiased Concentration (ICLR 2024 · similarity 0.84)
- SAS: Simulated Attention Score (NeurIPS 2025 · similarity 0.84)
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.
