Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
ContrastNet: A Contrastive Learning Framework for Few-Shot Text Classification
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 supervised contrastive text representation learning with task-level and instance-level regularization | few-shot text classification · Intent datasets average, 5-way 1-shot | accuracy: 89.13±1.13 -> 90.06±1.02 | Helped |
| Add supervised contrastive text representation learning with task-level and instance-level regularization | few-shot text classification · Intent datasets average, 5-way 5-shot | accuracy: 94.92±0.57 -> 95.29±0.53 | Helped |
| Add supervised contrastive text representation learning with task-level and instance-level regularization | few-shot text classification · News/review datasets average, 5-way 1-shot | accuracy: 63.9 -> 71.84±2.81 | Helped |
| Add supervised contrastive text representation learning with task-level and instance-level regularization | few-shot text classification · News/review datasets average, 5-way 5-shot | accuracy: 81.4 -> 81.85±2.03 | Helped |
| Add task-level and instance-level unsupervised contrastive regularization to supervised contrastive learning (ablation) | ContrastNet · Intent datasets average, 5-way 1-shot | accuracy: 86.99±1.57 -> 90.06±1.02 | Helped |
| Add task-level and instance-level unsupervised contrastive regularization to supervised contrastive learning (ablation) | ContrastNet · Intent datasets average, 5-way 5-shot | accuracy: 94.68±0.74 -> 95.29±0.53 | Helped |
| Add task-level and instance-level unsupervised contrastive regularization to supervised contrastive learning (ablation) | ContrastNet · News/review datasets average, 5-way 1-shot | accuracy: 70.45±3.28 -> 71.84±2.81 | Helped |
| Add task-level and instance-level unsupervised contrastive regularization to supervised contrastive learning (ablation) | ContrastNet · News/review datasets average, 5-way 5-shot | accuracy: 80.35±3.32 -> 81.85±2.03 | Helped |
| Add instance-level unsupervised contrastive regularization (ablation) | ContrastNet with task-level regularization · Intent datasets average, 5-way 1-shot | accuracy: 89.50±1.30 -> 90.06±1.02 | Helped |
| Add instance-level unsupervised contrastive regularization (ablation) | ContrastNet with task-level regularization · Intent datasets average, 5-way 5-shot | accuracy: 94.65±0.64 -> 95.29±0.53 | Helped |
| Add instance-level unsupervised contrastive regularization (ablation) | ContrastNet with task-level regularization · News/review datasets average, 5-way 1-shot | accuracy: 70.91±3.00 -> 71.84±2.81 | Helped |
| Add instance-level unsupervised contrastive regularization (ablation) | ContrastNet with task-level regularization · News/review datasets average, 5-way 5-shot | accuracy: 80.73±2.79 -> 81.85±2.03 | Helped |
Closest papers by meaning
- A Simple Graph Contrastive Learning Framework for Short Text Classification (AAAI 2025 · similarity 0.90)
- Improved Graph Contrastive Learning for Short Text Classification (AAAI 2024 · similarity 0.89)
- Contrastive Learning with Simplicial Convolutional Networks for Short-Text Classification (ICML 2025 · similarity 0.89)
- Boosting Short Text Classification with Multi-Source Information Exploration and Dual-Level Contrastive Learning (AAAI 2025 · similarity 0.87)
- Liberating Seen Classes: Boosting Few-Shot and Zero-Shot Text Classification via Anchor Generation and Classification Reframing (AAAI 2024 · similarity 0.87)
- Weakly-Supervised Contrastive Learning for Imprecise Class Labels (ICML 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.