Research map / Label-efficient and robust learning / Recognition and long-tail
Global Mixup: Eliminating Ambiguity with Clustering
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 |
|---|---|---|---|
| Add Global Mixup to CNNsen | CNNsen · YELP | accuracy (%): 92.5\pm 0.22 -> 93.4 \pm 0.13 | Helped |
| Add Global Mixup to CNNsen | CNNsen · SST-1 | accuracy (%): 36.5 \pm 0.45 -> 38.5 \pm 0.20 | Helped |
| Add Global Mixup to CNNsen | CNNsen · SST-2 | accuracy (%): 79.3 \pm 0.49 -> 80.1 \pm 0.23 | Helped |
| Add Global Mixup to CNNsen | CNNsen · TREC | accuracy (%): 97.8 \pm 0.32 -> 98.2\pm 0.19 | Helped |
| Add Global Mixup to CNNsen | CNNsen · SUBJ | accuracy (%): 91.9 \pm 0.21 -> 92.8 \pm 0.23 | Helped |
| Add Global Mixup to LSTMsen | LSTMsen · YELP | accuracy (%): 92.8 \pm 0. 22 -> 94.0 \pm 0.16 | Helped |
| Add Global Mixup to LSTMsen | LSTMsen · SST-1 | accuracy (%): 38.4 \pm 0. 74 -> 39.9 \pm 0.46 | Helped |
| Add Global Mixup to LSTMsen | LSTMsen · SST-2 | accuracy (%): 81.1 \pm 0.66 -> 81.6 \pm 0.31 | Helped |
| Add Global Mixup to LSTMsen | LSTMsen · TREC | accuracy (%): 98.1 \pm 0. 63 -> 98.6 \pm 0.35 | Helped |
| Add Global Mixup to LSTMsen | LSTMsen · SUBJ | accuracy (%): 92.6 \pm 0. 42 -> 93.1 \pm 0.39 | Helped |
| Add Global Mixup to BERT | BERT · YELP | accuracy (%): 97.0 \pm 0.13 -> 97.1 \pm 0.15 | Helped |
| Add Global Mixup to BERT | BERT · SST-1 | accuracy (%): 52.0 \pm 0. 64 -> 52.8 \pm 0.32 | Helped |
Closest papers by meaning
- C-Mixup: Improving Generalization in Regression (NeurIPS 2022 · similarity 0.91)
- Adversarial AutoMixup (ICLR 2024 · similarity 0.87)
- Catch-Up Mix: Catch-Up Class for Struggling Filters in CNN (AAAI 2024 · similarity 0.86)
- Pushing Boundaries: Mixup's Influence on Neural Collapse (ICLR 2024 · similarity 0.86)
- For Better or For Worse? Learning Minimum Variance Features With Label Augmentation (ICLR 2025 · similarity 0.86)
- Tailoring Mixup to Data for Calibration (ICLR 2025 · similarity 0.86)
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.