论文标题
共同培训改善了大语言模型的基于迅速的学习
Co-training Improves Prompt-based Learning for Large Language Models
论文作者
论文摘要
我们证明，共同训练（Blum＆Mitchell，1998）可以通过使用未标记的数据来提高基于及时的学习的性能。虽然提示已成为几次射击和零照片学习的有希望的范式，但与标准监督设置相比，它通常是脆弱的，并且需要更大的模型。我们发现，共同训练可以改善原始及时模型，同时学习一个较小的，下游特定于任务的模型。在我们仅部分访问及时模型的情况下（例如，GPT-3的输出概率（Brown等，2020）），我们会在提示输出上学习校准模型。当我们可以完全访问及时模型的梯度，但完整的填充仍然过于昂贵（例如，T0（Sanh等，2021））时，我们将学习一组软提示向量，以迭代更新及时模型。我们发现，以这种方式训练的模型可以显着提高在挑战性数据集上的性能，在这些数据集中，目前基于及时的学习和完全监督模型之间存在较大差距。
We demonstrate that co-training (Blum & Mitchell, 1998) can improve the performance of prompt-based learning by using unlabeled data. While prompting has emerged as a promising paradigm for few-shot and zero-shot learning, it is often brittle and requires much larger models compared to the standard supervised setup. We find that co-training makes it possible to improve the original prompt model and at the same time learn a smaller, downstream task-specific model. In the case where we only have partial access to a prompt model (e.g., output probabilities from GPT-3 (Brown et al., 2020)) we learn a calibration model over the prompt outputs. When we have full access to the prompt model's gradients but full finetuning remains prohibitively expensive (e.g., T0 (Sanh et al., 2021)), we learn a set of soft prompt continuous vectors to iteratively update the prompt model. We find that models trained in this manner can significantly improve performance on challenging datasets where there is currently a large gap between prompt-based learning and fully-supervised models.
