Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Co-training Improves Prompt-based Learning for Large Language Models
Hunter Lang, Monica N Agrawal, Yoon Kim, David Sontag
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 co-training to the GPT-3 label model (ablation) | GPT-3 label model · RTE | accuracy: 62.8 -> 64.9 | Improved |
| Add co-training to the GPT-3 label model (ablation) | GPT-3 label model · CB | accuracy: 76.8 -> 83.5 | Improved |
| Add co-training to the GPT-3 label model (ablation) | GPT-3 label model · TREC | accuracy: 77.2 -> 78.3 | Improved |
| Add co-training to the T0-3B soft prompt (ablation) | T0-3B · RTE | accuracy: 68.9 -> 87.0 | Improved |
| Add co-training to the T0-3B soft prompt (ablation) | T0-3B · CB | accuracy: 58.9 -> 67.9 | Improved |
| Add co-training to the T0-3B soft prompt (ablation) | T0-3B · BoolQ | accuracy: 56.4 -> 49.1 | Hurt |
| Use GPT-3 prompt outputs to co-train DeBERTa-large | DeBERTa-large · RTE | accuracy | Improved |
| Use GPT-3 prompt outputs to co-train DeBERTa-large | DeBERTa-large · CB | accuracy | Improved |
| Use GPT-3 prompt outputs to co-train DeBERTa-large | DeBERTa-large · TREC | accuracy | Improved |
| Use T0-3B prompt outputs to train DeBERTa-large | DeBERTa-large · BoolQ | accuracy: 56.4 -> 48.9 | Hurt |
| Use T0-3B prompt outputs to train DeBERTa-large | DeBERTa-large · CB | accuracy: 58.9 -> 67.9 | Mixed |
| Use a cut statistic to select confident data for the T0 view | T0-3B co-training · CB | pseudo-label accuracy: 89% | No clear effect |
Closest papers by meaning
- Learning to Learn Better Visual Prompts (AAAI 2024 · similarity 0.89)
- Learning to Prompt with Text Only Supervision for Vision-Language Models (AAAI 2025 · similarity 0.89)
- PromptKD: Unsupervised Prompt Distillation for Vision-Language Models (CVPR 2024 · similarity 0.88)
- SEC-Prompt:SEmantic Complementary Prompting for Few-Shot Class-Incremental Learning (CVPR 2025 · similarity 0.88)
- Weighted Multi-Prompt Learning with Description-free Large Language Model Distillation (ICLR 2025 · similarity 0.88)
- Batch Calibration: Rethinking Calibration for In-Context Learning and Prompt Engineering (ICLR 2024 · similarity 0.88)
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.
